Hacker News

OpenAI may have called GPT-5.2 its "most advanced frontier model for professional work," but tests conducted by the <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:1;pos:1" class="no-affiliate-link" href="https://www.theguardian.com/technology/2026/jan/24/latest-chatgpt-model-uses-elon-musks-grokipedia-as-source-tests-reveal">Guardian</a> cast doubt on its credibility. According to the report, OpenAI's GPT-5.2 model cited <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:2;pos:1" class="no-affiliate-link" href="https://www.engadget.com/social-media/xs-grokipedia-is-online-after-it-briefly-crashed-out-231108836.html">Grokipedia</a>, the online encyclopedia powered by xAI, when it came to specific, but controversial topics related to Iran or the Holocaust. As seen in the Guardian's report, ChatGPT used Grokipedia as a source for claims about the Iranian government being tied to telecommunications company MTN-Irancell and questions related to Richard Evans, a British historian who served as an expert witness during a libel trial for Holocaust denier David Irving. However, the Guardian noted ChatGPT didn't use Grokipedia when it came to a prompt asking about media bias against Donald Trump and other controversial topics. OpenAI released the <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:3;pos:1" class="no-affiliate-link" href="https://www.engadget.com/ai/openai-releases-gpt-52-to-take-on-google-and-anthropic-185029007.html">GPT-5.2 model</a> in December to better perform at professional use, like creating spreadsheets or handling complex tasks. Grokipedia preceded GPT-5.2's release, but ran into some controversy when it was seen <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:4;pos:1" class="no-affiliate-link" href="https://www.engadget.com/ai/authorities-probe-holocaust-denial-responses-from-xs-grok-155835701.html">including citations</a> to neo-Nazi forums. A study done by US researchers also showed that the AI-generated encyclopedia <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:5;pos:1" class="no-affiliate-link" href="https://www.france24.com/en/live-news/20251114-musk-s-grokipedia-leans-on-questionable-sources-study-says">cited</a> "questionable" and "problematic" sources. In response to the Guardian report, OpenAI told the outlet that its GPT-5.2 model searches the web for a "broad range of publicly available sources and viewpoints," but applies "safety filters to reduce the risk of surfacing links associated with high-severity harms."This article originally appeared on Engadget at https://www.engadget.com/ai/report-reveals-that-openais-gpt-52-model-cites-grokipedia-192532977.html?src=rss

Jan 24, 2026

Google says it's working to fix Gmail issue that's led to flooded inboxes and increased spam warnings

Jan 24, 2026

US Congress members call for 'thorough review' of EA's $55 billion sale

Before <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:1;pos:1" class="no-affiliate-link" href="https://www.engadget.com/gaming/ea-delays-battlefield-6-season-2-to-february-17-180306743.html">Electronic Arts</a> goes private in a groundbreaking sale, some US lawmakers are pleading for some federal oversight. Democratic members of the US Congress, as part of the <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:2;pos:1" class="no-affiliate-link" href="https://laborcaucus.house.gov/sites/evo-subsites/laborcaucus.house.gov/files/evo-media-document/letter-to-ftc-on-proposed-acquisition-of-electronic-arts-ea.pdf">Congressional Labor Caucus</a>, penned a letter asking the Federal Trade Commission to "thoroughly review" the $55 billion acquisition of EA. <a data-i13n="cpos:3;pos:1" href="https://www.engadget.com/ea-confirms-it-will-go-private-in-55-billion-acquisition-133841614.html">EA confirmed</a> the sale to the Public Investment Fund, or the sovereign wealth fund of Saudi Arabia, Silver Lake and Affinity Partners in September, but the deal is expected to close in the first quarter of 2027. Before the official change of ownership, the 46 House Democrats who signed the letter to the FTC are calling for more scrutiny into the impacts of the deal.  The letter noted some of the most consequential effects, including the worsening of an unstable industry, the potential for more layoffs and increased <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:4;pos:1" class="no-affiliate-link" href="https://www.engadget.com/ea-purchasing-glu-mobile-045832340.html">market dominance</a> for EA. "We respectfully urge the Commission to conduct a thorough investigation into the labor market consequences of this proposed acquisition, including EA’s existing wage-setting power, the likelihood of post-transaction layoffs, the degree of labor-market concentration in relevant geographic and occupational markets, and the role of cross-ownership in shaping labor outcomes," the letter read. The letter already earned support from the Communications Workers of America union, who also supported a petition from the <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:5;pos:1" class="no-affiliate-link" href="https://actionnetwork.org/petitions/make-ea-better">United Video Games</a> union. As spotted by <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:6;pos:1" class="no-affiliate-link" href="https://www.eurogamer.net/uvw-union-fed-trade-commission-letter-petition">Eurogamer</a>, the petition calls on regulators and elected officials to "scrutinize this deal and ensure that any path forward protects jobs and preserves creative freedom."This article originally appeared on Engadget at https://www.engadget.com/gaming/us-congress-members-call-for-thorough-review-of-eas-55-billion-sale-175851429.html?src=rss

Jan 24, 2026

NTSB will investigate why Waymo's robotaxis are illegally passing school buses

<a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:1;pos:1" class="no-affiliate-link" href="https://www.engadget.com/transportation/evs/waymo-is-now-available-in-miami-162515070.html">Waymo</a> has caught the attention of the National Transportation Safety Board as the federal agency launched an official investigation into the company for its robotaxis improperly passing school buses in Austin, Texas. The NTSB <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:2;pos:1" class="no-affiliate-link" href="https://x.com/NTSB_Newsroom/status/2014817506477703198">said on X</a> that it would "examine the interaction between Waymo vehicles and school buses stopped for loading and unloading students." The latest federal probe stems from a preliminary evaluation by the National Highway Traffic Safety Administration that looked into how Waymo reacts to stopped school buses in the Texas city. That report led to Waymo's <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:3;pos:1" class="no-affiliate-link" href="https://www.engadget.com/transportation/waymos-robotaxi-fleet-is-being-recalled-again-this-time-for-failing-to-stop-for-school-buses-190222243.html">voluntary software recall</a> in December. However, the school district said <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:4;pos:1" class="no-affiliate-link" href="https://static.nhtsa.gov/odi/inv/2025/INOT-PE25013-30888P1.pdf">in a memo</a> that the robotaxis were seen repeating the same offense days after the software update. As for the NTSB investigation, an agency spokesperson told the <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:5;pos:1" class="no-affiliate-link" href="https://www.statesman.com/business/technology/article/ntsb-waymo-investigation-austin-school-bus-21312004.php">Austin American-Statesman</a> that its "investigators will travel to Austin to gather information on a series of incidents in which the automated vehicles failed to stop for loading or unloading students." According to an NTSB spokesperson, a preliminary report will be out within 30 days, but the final report will take anywhere between 12 and 24 months. In response, Mauricio Peña, chief safety officer for Waymo, said in a statement to <a data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:6;pos:1" class="no-affiliate-link" href="https://www.cbsnews.com/news/ntsb-investigation-waymo-robotaxis-passing-school-buses-austin-texas/">multiple news outlets</a> that "there have been no collisions in the events in question, and we are confident that our safety performance around school buses is superior to human drivers," adding that the investigation will be "an opportunity to provide the NTSB with transparent insights into our safety-first approach."This article originally appeared on Engadget at https://www.engadget.com/transportation/ntsb-will-investigate-why-waymos-robotaxis-are-illegally-passing-school-buses-160943613.html?src=rss

Jan 24, 2026

How to use Google Photos' new Me Meme feature

Google has started <a target="_blank" class="link" href="https://www.engadget.com/ai/google-photos-can-now-turn-you-into-a-meme-213930935.html" data-i13n="cpos:1;pos:1">rolling out</a> a new feature for its Photos app that can turn your images into memes. The feature, called Me Meme, uses Google Gemini to take meme templates and recreate them with the photo you use from within the app. It’s still in its experimental stages and will only be available for users in the US in the English language. In addition, it seems to be rolling out for Android users only at the moment, and Google warns that the generated images may not match the original photo at times. If you do get access to the feature soon, check out the instructions below on how to use it. <ol><li>Open your Photos app. Go to the Create tab at the bottom and find the Me Meme option. If it’s not showing up even though you’re an Android user in the US, you’ll have to wait for it as it continues rolling out. Google told <a target="_blank" class="link" href="https://techcrunch.com/2026/01/23/google-photos-latest-feature-lets-you-meme-yourself/" data-i13n="cpos:2;pos:1">TechCrunch</a> that it will reach iOS users over the coming weeks. </li><li>You’ll see onscreen instructions the first time you use it. The next time you access the feature, you’ll go straight to the meme creation process. </li><li>You can browse and choose one of the preset templates Google provides, but you can also upload a meme or any other image as a template. </li><li>You then have to upload a photo that you want to insert into the meme. Google advises using a selfie or a photo where your face is clearly visible. </li><li>After you tap Generate, you can save the meme if you’re satisfied with the result or share it directly to social media platforms online. You can also tap Regenerate to see a different output. </li></ol><a target="_blank" class="link" href="https://x.com/AssembleDebug/status/2014523215117484222" data-i13n="cpos:3;pos:1">Android Authority</a> spotted the feature in October 2025 and was able to try it out before its limited release. You can see an example of Me Meme’s output below.<div><blockquote class="twitter-tweet">The "Me meme" feature which I spotted back in October 2025 is finally announced by Google. This is an experimental feature which will be available first for users in US only 😐 Here is the announcement forum posts- <a href="https://t.co/7P2JgJhoBk">https://t.co/7P2JgJhoBk</a> <a href="https://t.co/E60prcqcie">https://t.co/E60prcqcie</a> <a href="https://t.co/sFICxzVIPU">pic.twitter.com/sFICxzVIPU</a>— AssembleDebug (Shiv) (@AssembleDebug) <a href="https://twitter.com/AssembleDebug/status/2014523215117484222?ref_src=twsrc%5Etfw">January 23, 2026</a></blockquote> </div>This article originally appeared on Engadget at https://www.engadget.com/ai/how-to-use-google-photos-new-me-meme-feature-140000157.html?src=rss

Jan 24, 2026

How to use Workout Buddy with Apple Watch and iOS 26

Apple’s iOS 26 and watchOS 26 introduced a new fitness companion called Workout Buddy. This feature uses Apple Intelligence to provide spoken feedback during workouts and give motivation based on your activity history. Workout Buddy analyzes your pace, heart rate, distance and other metrics to deliver real-time encouragement and performance insights directly through connected Bluetooth headphones. It works in conjunction with the Workout app on <a target="_blank" class="link" href="https://www.engadget.com/wearables/best-apple-watch-160005462.html" data-i13n="cpos:1;pos:1">Apple Watch</a> and is partially controlled through the Fitness app on <a target="_blank" class="link" href="https://www.engadget.com/mobile/smartphones/best-iphone-160012979.html" data-i13n="cpos:2;pos:1">iPhone</a>. This guide walks you through everything needed to set up and use Workout Buddy effectively during workouts.<h2 id="jump-link-what-workout-buddy-does">What Workout Buddy does</h2>It’s important to note that Workout Buddy is not a full coaching program. Instead, it adds to your workout with spoken cues that reflect how your session is going. Workout Buddy can remind you of your weekly activity totals, alert you to personal bests or performance milestones and provide an overview when you’re finished. It is designed to feel like a supportive training partner rather than a strict coach.The feature operates in English by default and uses a text-to-speech model trained on voices from Apple Fitness+ trainers. It is available for a subset of workout types, including running, walking, cycling, high-intensity interval training (HIIT) and strength training. It requires on-device Apple Intelligence, which means you’ll need to keep one of the latest iPhones running updated software nearby during workouts. Supported models include iPhone 15 Pro, iPhone 15 Pro Max and any iPhone 16 model. You’ll also need an Apple Watch running watchOS 26.  <h2 id="jump-link-requirements-before-you-begin">Requirements before you begin</h2>Before Workout Buddy appears in your Fitness app or Workout app you must ensure a few things are in place. First, your Apple Watch must be running watchOS 26 or later and paired to an iPhone with iOS 26 installed. Second, your iPhone must be capable of on-device Apple Intelligence, meaning you must own one of the supported iPhone models we mentioned above and have Apple Intelligence enabled in the phone’s settings.You’ll also need Bluetooth headphones paired with either your iPhone or your Apple Watch. Workout Buddy’s audio feedback cannot play through the watch speaker so headphones are essential. Lastly, your device language must be set to English, at least initially. If any of these things are missing, the option to enable Workout Buddy may not appear.<h2 id="jump-link-how-to-turn-on-workout-buddy-from-iphone">How to turn on Workout Buddy from iPhone</h2>While much of the interaction with Workout Buddy happens on Apple Watch during workouts, you can enable it and choose voice options from the Fitness app on iPhone.Open the Fitness app on your iPhone and tap the Workout tab at the bottom. Scroll through the list of workout types until you find one you plan to use with Workout Buddy. Tap the waveform bubble icon associated with that workout. This will bring up settings where you can turn on Workout Buddy. Flip the toggle to enable it and choose a voice from the available options. Once you have selected a voice, close that screen and your choice is saved. When you start this workout type on Apple Watch, Workout Buddy will activate.Enabling Workout Buddy for a workout type on iPhone means you do not need to toggle it on separately on Apple Watch each time for that specific workout. However, you may still adjust it from the watch interface for more granular control.<h2 id="jump-link-how-to-turn-on-workout-buddy-on-apple-watch">How to turn on Workout Buddy on Apple Watch</h2>To use Workout Buddy during a session, open the Workout app on your Apple Watch. Turn the Digital Crown to scroll through and select the workout you want to do, such as Outdoor Run, Outdoor Walk, Outdoor Cycle, HIIT or Strength Training. If you want to see all available workouts, tap the Add button at the bottom.Once the workout type is selected, look for the Alerts button on screen. Tap Alerts then scroll until you see Workout Buddy. Tap Workout Buddy and flip the switch to on. You will then be asked to choose a voice if one is not already selected on your iPhone. After selecting the voice, return to the previous screen and tap Start. Workout Buddy will begin working as soon as the workout does.<h2 id="jump-link-using-workout-buddy-during-a-workout">Using Workout Buddy during a workout</h2>Once you start an exercise on your Watch or iPhone, Workout Buddy will speak to you through your connected headphones. The feedback is designed to be encouraging and relevant to your pace, performance or milestones. It may mention your current progress toward activity goals, pace, splits, personal bests or other highlights from your fitness data. At the end of your session Workout Buddy will offer a summary of key metrics like duration distance and calorie burn.While a workout is active, you can temporarily mute the audio if you need silence. On Apple Watch during the session, swipe right to reveal controls then tap Mute. This pauses Workout Buddy’s spoken commentary without disabling the feature entirely.<h2 id="jump-link-customizing-and-managing-workout-buddy-settings">Customizing and managing Workout Buddy settings</h2>Workout Buddy is enabled on a per-workout-type basis. If you prefer voice feedback for running but silence for strength training, you can enable it for one and leave it off for the other. The Fitness app on iPhone allows you to set a default voice preference for each workout type. On Apple Watch you can quickly toggle the feature on or off before starting a session.If Workout Buddy does not appear as an option for a particular workout type, you may need to check compatibility. Apple’s documentation indicates that only certain types* are supported initially and that the option will not appear for unsupported workouts.*Apple Watch SE (2nd generation), Apple Watch SE 3, Apple Watch Series 6, Apple Watch Series 7, Apple Watch Series 8, Apple Watch Series 9, Apple Watch Series 10, Apple Watch Series 11, Apple Watch Ultra, Apple Watch Ultra 2, Apple Watch Ultra 3<h2 id="jump-link-troubleshooting-common-issues">Troubleshooting common issues</h2>If Workout Buddy fails to activate make sure your devices meet the requirements outlined above. Confirm that your iPhone with Apple Intelligence is nearby and that Bluetooth headphones are connected. If audio feedback is missing, ensure headphones are paired correctly and that the language is set to English. Some users have reported that if the headphones are paired only to the Watch rather than the iPhone, it can interfere with feedback. Switching to the iPhone often resolves that issue.For workout types where Workout Buddy previously worked but suddenly does not appear, you may try toggling the feature off and on again in the Fitness app or rebooting both devices. In rare cases removing and re-adding the workout type on Apple Watch can refresh the settings.This article originally appeared on Engadget at https://www.engadget.com/wearables/how-to-use-workout-buddy-with-apple-watch-and-ios-26-130000922.html?src=rss

Jan 24, 2026

Engadget review recap: Valerion VisionMaster Max, Canon EOS R6 III and Samsung Bespoke Fridge

Our <a target="_blank" class="link" href="https://www.engadget.com/reviews/" data-i13n="cpos:1;pos:1">reviews team</a> is rested up after the CES grind and we’re back to business. This week we reviewed the latest devices from Valerion, Canon, Samsung and Sony, covering a wide range of product categories in the process. Grab yourself a beverage and cozy up with all the reviews you might’ve missed. <h2 id="jump-link-valerion-visionmaster-max">Valerion VisionMaster Max</h2> <core-commerce data-type="product-list" id="d449ef81-f1d7-4721-a3dc-c189a188e3e2" data-original-url="https://www.amazon.com/Valerion-VisionMaster-Max-Projector-Speckle-Reduction/dp/B0FQNXHVGS/"></core-commerce> The weeks leading up to the Super Bowl are a popular time to upgrade home theater setups before the big game. One way to do that is <a target="_blank" class="link" href="https://www.engadget.com/home/home-theater/valerion-visionmaster-max-projector-review-near-perfect-image-quality-comes-at-a-price-140045939.html" data-i13n="cpos:2;pos:1">Valerion’s VisionMaster Max</a> — if you’re willing to splurge. “The Valerion VisionMaster Max is a highly capable indoor projector that offers the best image quality I’ve seen thanks to the dynamic iris and Enhanced Black Level features,” contributing reporter Steve Dent discovered. “However, it’s also a lot more expensive than rival models at $5,000. It’s pretty hard to justify that extra money, unless you’re really fussy about picture accuracy and gaming performance, or need other features like the 1Gbps ethernet port.” <h2 id="jump-link-canon-eos-r6-iii">Canon EOS R6 III</h2> <core-commerce data-type="product-list" id="d3332c75-3ae0-4d3d-a029-317ed2726286" data-original-url="https://www.amazon.com/Canon-EOS-Mark-III-Body/dp/B0FZVVLR9D/ref=sr_1_1ZVVLR9D/ref=sr_1_1"></core-commerce> If you’re primarily shooting video these days, contributing reporter Steve Dent just wrapped up testing with the <a target="_blank" class="link" href="https://www.engadget.com/cameras/canon-eos-r6-iii-review-a-video-powerhouse-thats-also-great-for-photography-153000494.html" data-i13n="cpos:3;pos:1">Canon EOS R6 III</a>. While he said it’s great for video, the photo quality isn’t the absolute best you can get right now. “If it’s mainly photography you’re interested in, though, Sony’s $2,900 A7 V is a better choice,” he said. “It matches the R6 III’s resolution, price and autofocus but offers superior dynamic range and lower rolling shutter distortion.” <h2 id="jump-link-samsung-bespoke-fridge-with-ai">Samsung Bespoke Fridge with AI</h2> <core-commerce data-type="product-list" id="c2559660-5426-4082-9350-0af25e275527" data-original-url="https://www.samsung.com/us/home-appliances/refrigerators/bespoke/bespoke-23-cu-ft-counter-depth-4-door-french-door-refrigerator-with-ai-family-hub-ai-vision-inside-in-white-glass-rf90f23aewaa/"></core-commerce> We don’t often review appliances here at Engadget, but <a target="_blank" class="link" href="https://www.engadget.com/home/kitchen-tech/samsung-bespoke-fridge-with-ai-review-all-the-bells-and-whistles-140000099.html" data-i13n="cpos:4;pos:1">Samsung’s AI-infused refrigerator</a> caught the attention of senior reporter Sam Rutherford. You may not think you need a 32-inch display in your kitchen or cameras in your fridge, but Sam argues once you try them you may never go back. “You may not agree or want to pay this much for features like this just yet, but Samsung's Bespoke AI 4-Door French Door refrigerator has convinced me there's a happy home for smarter iceboxes,” he concluded. <h2 id="jump-link-sony-linkbuds-clip">Sony LinkBuds Clip</h2> <core-commerce data-type="product-list" id="22bfaf8f-59aa-40cf-8290-5b3fffb85ec2" data-original-url="https://www.amazon.com/Sony-Comfortable-Lightweight-Awareness-Resistant/dp/B0G4SKRJG7?th=1"></core-commerce> Clip-on earbuds have been constantly popping up over the last few years. Many of the key players in headphones have tried their hand at the open-wear form factor with Sony being the latest to jump on board. While the company embraces the inherent perks of these designs, the <a target="_blank" class="link" href="https://www.engadget.com/audio/headphones/sony-linkbuds-clip-review-open-fit-benefits-arent-enough-to-stand-out-160000140.html" data-i13n="cpos:5;pos:1">LinkBuds Clip</a> doesn’t do enough to rise above the competition. “Subpar bass performance and the omission of some of Sony’s more attractive features (and even some basic ones) mean the company hasn’t done enough to distinguish the Clip from the competition in an obvious way,” I wrote.This article originally appeared on Engadget at https://www.engadget.com/engadget-review-recap-valerion-visionmaster-max-canon-eos-r6-iii-and-samsung-bespoke-fridge-123000849.html?src=rss

Jan 24, 2026

More Cult of the Lamb, a World War II computer mystery and other new indie games worth checking out

Welcome to our latest roundup of what's going on in the indie game space. It’s been a very busy week of fun game releases (next week will be too!), so let’s get into some of them after a quick reflection on gaming while traveling.I love my Steam Deck. I really truly do. It’s a fantastic machine. And yet when I brought it with me on a five-week trip over the holidays, I used it for barely an hour the entire time. That doesn’t really justify the space and weight it takes up in my bag. The same holds true for my Nintendo Switch 2 and PlayStation Portal. I’ll be leaving them all at home next time. I’ll take a small controller (probably <a target="_blank" class="link" href="https://www.engadget.com/gaming/another-look-at-ohsnaps-mcon-a-magnetic-game-controller-for-your-phone-042424092.html" data-i13n="cpos:1;pos:1">OhSnap's MCON</a>) so I can play the odd game on my laptop or phone. I’ll bring my Playdate as well. I adore that little yellow console, and I haven’t spent nearly enough time using it. I have a lot of neat-looking indie games to catch up on there (<a target="_blank" class="link" href="https://play.date/games/diora/" data-i13n="cpos:2;pos:1">Diora</a> looks particularly interesting). I’m more likely to play something on a flight if it’s on a device I can pull out from my pocket. I just wish Panic had put a backlight in the Playdate. <h2 id="jump-link-new-releases">New releases</h2><div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/9YH21SghaUU?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Massive Monster and publisher Devolver Digital gave <a target="_blank" class="link" href="https://www.engadget.com/cult-of-the-lamb-is-finally-getting-a-co-op-mode-but-its-local-only-180715026.html" data-i13n="cpos:3;pos:1">Cult of the Lamb</a> fans a whole bunch of reasons to jump back into the game this week with the arrival of the <a target="_blank" class="link" href="https://www.engadget.com/gaming/cult-of-the-lambs-next-dlc-is-woolhaven-out-in-early-2026-225129643.html" data-i13n="cpos:4;pos:1">Woolhaven expansion</a>. You'll need to get close to the end of the base game (though you don't need to beat the final boss) before you can experience what the DLC has to offer. It includes weather effects, a new mountain area with a pair of fresh dungeons, <a target="_blank" class="link" href="https://www.engadget.com/this-is-how-the-new-ranching-system-will-work-in-cult-of-the-lamb-woolhaven-194324041.html" data-i13n="cpos:5;pos:1">a ranching system</a> (which allows you to raise animals as pets or for food) and much more.Folks who dig Cult of the Lamb seem to get really into the game. After a few years of free updates from Massive Monster, Woolhaven is a major expansion that's similar in scope to the base game, so it should keep fans busy for quite a while if they want to try everything. The DLC is out now on <a target="_blank" class="link" href="https://store.steampowered.com/app/3840050/Cult_of_the_Lamb_Woolhaven/" data-i13n="cpos:6;pos:1">PC</a>, Nintendo Switch, PS4, PS5, Xbox Series X/S and Xbox One. It costs $17 if you already have the base game.<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/NEIyvjVQCAE?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Inkle, the studio behind Overboard! and the wonderful A Highland Song is back with TR-49, a puzzle game with a World War II computer at its core. Over five decades, an array of books, letters and journals were fed into the machine as part of an effort to "crack the code of reality." But now you're tasked with finding and destroying one specific book before a timer runs out.It all seems rather mysterious. To give you a sense of what's going on here, Inkle says it drew inspiration from narrative deduction games like The Roottrees are Dead, The Return of the Obra Dinn and Her Story, as well as audio dramas. I can't see myself having the time or patience to figure out the enigma of this particular machine (I imagine many players will need a notebook for this one), but I'm intrigued enough to at least watch a Let's Play at some point. You can experience this mystery for yourself on <a target="_blank" class="link" href="https://store.steampowered.com/app/3838370/TR49/" data-i13n="cpos:7;pos:1">Steam</a> (normally $7, but there's a 10 percent launch discount until January 28).<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/cafPCloPgtc?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Another week, another Metroidvania, but MIO: Memories in Orbit has a more striking art style than most. This game from Douze Dixièmes and publisher Focus Entertainment debuted to positive reviews this week. After the AI caretakers on a forgotten spaceship stop working, you'll have to help the robot MIO explore the spacecraft "to revive its lost memories" and save it from doom. Naturally, you'll discover some new abilities along the way, including a grappling hook and air gliding.MIO: Memories in Orbit is available on <a target="_blank" class="link" href="https://store.playstation.com/en-us/concept/10010037" data-i13n="cpos:8;pos:1">PS5</a>, <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=5f41950e-8ca3-4481-8466-a22b28b80e32&siteId=us-engadget&pageId=1p-autolink&contentUuid=86d78fd7-706f-45bc-8ccc-e3d69ab5f680&featureId=text-link&merchantName=Xbox&linkText=Xbox+Series+X%2FS&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy54Ym94LmNvbS9lbi1VUy9nYW1lcy9zdG9yZS9taW8tbWVtb3JpZXMtaW4tb3JiaXQvOW5wODZrNTdzcW0zIiwiY29udGVudFV1aWQiOiI4NmQ3OGZkNy03MDZmLTQ1YmMtOGNjYy1lM2Q2OWFiNWY2ODAiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3Lnhib3guY29tL2VuLVVTL2dhbWVzL3N0b3JlL21pby1tZW1vcmllcy1pbi1vcmJpdC85bnA4Nms1N3NxbTMifQ&signature=AQAAAfWbXK45jmpjFawfpYsSR87Sl8iitEZNT22MkcoQFtF5&gcReferrer=https%3A%2F%2Fwww.xbox.com%2Fen-US%2Fgames%2Fstore%2Fmio-memories-in-orbit%2F9np86k57sqm3" data-i13n="elm:affiliate_link;sellerN:Xbox;elmt:;cpos:9;pos:1" data-original-link="https://www.xbox.com/en-US/games/store/mio-memories-in-orbit/9np86k57sqm3">Xbox Series X/S</a>, <a target="_blank" class="link" href="https://www.nintendo.com/us/store/products/mio-memories-in-orbit-switch" data-i13n="cpos:10;pos:1">Nintendo Switch</a>, <a target="_blank" class="link" href="https://www.nintendo.com/us/store/products/mio-memories-in-orbit-switch-2" data-i13n="cpos:11;pos:1">Switch 2</a>, <a target="_blank" class="link" href="https://store.steampowered.com/app/1672810/MIO_Memories_in_Orbit/" data-i13n="cpos:12;pos:1">Steam</a>, the <a target="_blank" class="link" href="https://store.epicgames.com/en-US/p/mio-memories-in-orbit-e46c84" data-i13n="cpos:13;pos:1">Epic Games Store</a> and the Microsoft Store for $20 (with a 10 percent launch discount on some platforms). You can also check it out via Game Pass Ultimate and PC Game Pass.<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/ESUKobGZcH8?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Perfect Tides: Station to Station is a pixel-art point-and-click narrative adventure and a sequel to Three Bees' Perfect Tides. Over the course of an in-game year, you'll help 18-year-old Mara explore the big city and find her place in the world. The early 2000s vibes of the trailer are immaculate, so I'm going to have to try to play this at some point. Perfect Tides: Station to Station is out now on <a target="_blank" class="link" href="https://store.steampowered.com/app/2088810/Perfect_Tides_Station_to_Station/" data-i13n="cpos:14;pos:1">Steam</a> for PC and Mac for $20.<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/pKO6MGVoXMI?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Cozy cafe sim Tailside debuted in early access on <a target="_blank" class="link" href="https://store.steampowered.com/app/2922570/Tailside_Cozy_Cafe_Sim/" data-i13n="cpos:15;pos:1">Steam</a> (normally $11, but there's a 10 percent discount until January 28) this week. I'm bummed I haven't had a chance to try this one from Coffee Beans Dev yet, because it looks lovely. Along with serving snacks and drinks to your furry customers, you can learn more about the visitors to your cafe by reading stories about them in the newspaper (hopefully nice ones!). You can play at your own pace as you draw latte art and decorate your cafe. Eventually, you'll be able to visit other players' cafes and open a flower shop. Like I said, it looks lovely.<h2 id="jump-link-upcoming">Upcoming </h2><div><blockquote class="twitter-tweet">BALL x PIT’s first FREE major update, The Regal Update, launches on January 26th on all platforms! Adding: 🤠2 new characters - The Carouser & The Falconer 🏐8 new balls 🎯New passives ✨Plus a nice surprise you’re going to love! <a href="https://t.co/t0tYlwjOXk">pic.twitter.com/t0tYlwjOXk</a>— BALL x PIT 🏐 OUT NOW (@BALLxPIT) <a href="https://twitter.com/BALLxPIT/status/2013296280261234716?ref_src=twsrc%5Etfw">January 19, 2026</a></blockquote> </div>One of my favorite games of last year is getting <a target="_blank" class="link" href="https://www.engadget.com/gaming/ball-x-pit-is-evolving-with-a-string-of-free-updates-in-2026-181947946.html" data-i13n="cpos:16;pos:1">a free update</a> on January 26. Just when I thought I was out of Ball x Pit, Kenny Sun and friends (along with publisher Devolver) are pulling me right back in to check out two new characters, eight fresh balls with their own abilities, more passives and some kind of surprise. <a target="_blank" class="link" href="https://www.engadget.com/gaming/ball-x-pits-deeply-satisfying-grind-keeps-me-coming-back-for-more-171000754.html" data-i13n="cpos:17;pos:1">I adore Ball x Pit</a> and this update is going to be a drop everything and play immediately deal for me. The Regal Update will be available on all platforms: Steam, PS5, Xbox Series X/S, Nintendo Switch and Switch 2. Ball x Pit costs $15. It's available via Game Pass Ultimate and PC Game Pass too.<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/-Ael4oMaKQM?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>I'm interested to check out the demo for <a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/a-vampire-survivors-roguelike-deckbuilder-spinoff-is-coming-in-2026-183841289.html" data-i13n="cpos:18;pos:1">Vampire Crawlers</a>, a Vampire Survivors spin-off. Poncle will <a target="_blank" class="link" href="http://store.steampowered.com/app/3265700/Vampire_Crawlers_The_Turbo_Wildcard_from_Vampire_Survivors/" data-i13n="cpos:19;pos:1">release it</a> on February 23 at Steam Next Fest as well as on Xbox. Your progress will carry over into the full game, which will be on Game Pass on day one.As a rule, turn-based games aren't really my jam, and nor are roguelike deckbuilders. But I'm a big fan of Vampire Survivors, so I'm definitely willing to give this a shot. It helps that Vampire Crawlers seems to be fast-paced and that it draws from the chaotic visuals of the original game. Vampire Crawlers is coming to Steam, Xbox Series X/S, PS5, Nintendo Switch, iOS and Android later this year.<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/laq3IrWbsVE?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Speaking of interesting demos, one for Ratcheteer DX is available now on PC ahead of the full game's release on <a target="_blank" class="link" href="https://www.nintendo.com/en-ca/store/products/ratcheteer-dx-switch" data-i13n="cpos:20;pos:1">Switch</a>, <a target="_blank" class="link" href="https://store.steampowered.com/app/3166150/Ratcheteer_DX/" data-i13n="cpos:21;pos:1">Steam</a> (PC and Mac) and the Mac App Store on March 5. It'll normally cost $13, but a limited-time discount will reduce the price to $9.75. If the game's name sounds familiar, that's because this is a color version of the Playdate season one title Ratcheteer. This take on the pixel-art action-adventure has multiple visual filters, a "CD-quality stereo soundtrack" and support for more languages. Playdate creator Panic is publishing Ratcheteer DX, whose developers are Shaun Inman, Matthew Grimm and Charlie Davis.<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/d6SYjoGrpwY?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Let's wrap things up for this week with a cinematic trailer and release date for Aethus, a story-driven survival-crafting and base-building game from a solo developer at Pawsmonaut Games. It's coming to <a target="_blank" class="link" href="https://store.steampowered.com/app/2541530/AETHUS/" data-i13n="cpos:22;pos:1">Steam</a> on March 6.Aethus is a game about "carving out a future from the ruins of corporate greed" in a dystopian sci-fi world. As ex-mining engineer Maeve, you start with basic gear and a drone companion by your side. You'll explore what's left of a failed science expedition and abandoned facilities as you try to establish your own mining claim after leaving a company that drains planets of their resources. Of course, you'll upgrade your gear and build out your base as you progress. You can modify settings like the base's air supply, how quickly your hunger and thirst needs change and how much you can carry to fine tune the experience. This article originally appeared on Engadget at https://www.engadget.com/gaming/more-cult-of-the-lamb-a-world-war-ii-computer-mystery-and-other-new-indie-games-worth-checking-out-120000807.html?src=rss

Jan 23, 2026

Google Photos can now turn you into a meme

Jan 23, 2026

A rival smart glasses company is suing Meta over its Ray-Ban products

Meta is being sued by <a target="_blank" class="link" href="https://solosglasses.com/" data-i13n="cpos:1;pos:1">Solos</a>, a rival smart glasses maker, for infringing on its patents, <a target="_blank" class="link" href="https://www.bloomberg.com/news/articles/2026-01-23/meta-essilorluxottica-sued-in-smart-glasses-patent-case" data-i13n="cpos:2;pos:1">Bloomberg reports</a>. Solos is seeking "multiple billions of dollars" in damages and an injunction that could prevent Meta from selling its Ray-Ban Meta smart glasses as part of the lawsuit.Solos claims that Meta's <a target="_blank" class="link" href="https://www.engadget.com/ray-ban-meta-smart-glasses-review-instagram-worthy-shades-070010365.html" data-i13n="cpos:3;pos:1">Ray-Ban Meta Wayfarer Gen 1 smart glasses</a> violate multiple patents covering "core technologies in the field of smart eyewear." While less well known than Meta and its partner EssilorLuxottica, Solos sells multiple pairs of glasses with similar features to what Meta offers. For example, the company's <a target="_blank" class="link" href="https://solosglasses.com/" data-i13n="cpos:4;pos:1">AirGo A5 glasses</a> lets you control music playback and automatically translate speech into different languages, and integrates ChatGPT for answering questions and searching the web.Beyond the product similarities, Solos claims that Meta was able to copy its patents because Oakley (an EssilorLuxottica subsidiary) and Meta employees had insights into the company's products and road map. Solos says that in 2015, Oakley employees were introduced to the company's smart glasses tech, and were even given a pair of Solos glasses for testing in 2019. Solos also says that a MIT Sloan Fellow who researched the company's products and later became a product manager at Meta, brought knowledge of the company to her role. According to the logic of Solos' lawsuit, by the time Meta and EssilorLuxottica were selling their own smart glasses, "both sides had accumulated years of direct, senior-level and increasingly detailed knowledge of Solos' smart glasses technology."Engadget has asked both Meta and EssilorLuxottica to comment on Solos' claims. We'll update this article if we hear back.While fewer people own Ray-Ban Meta smart glasses than use Instagram, Meta considers the wearable one of its few hardware success stories. The company is so convinced it can make smart glasses happen that it <a target="_blank" class="link" href="https://www.engadget.com/ar-vr/meta-refocuses-on-ai-hardware-as-metaverse-layoffs-begin-145924706.html" data-i13n="cpos:5;pos:1">recently restructured its Reality Labs division</a> to focus on AI hardware like smart glasses and hopefully build on its success. This article originally appeared on Engadget at https://www.engadget.com/wearables/a-rival-smart-glasses-company-is-suing-meta-over-its-ray-ban-products-205000997.html?src=rss

Jan 23, 2026

Retro handheld maker Anbernic has a new gamepad with a screen and heart rate sensor

File this one under "Things that make you go, 'Hmmm…'" <a target="_blank" class="link" href="https://www.engadget.com/gaming/you-can-now-buy-anbernics-budget-ds-clone-but-dont-get-your-hopes-up-for-3ds-emulation-183140820.html" data-i13n="cpos:1;pos:1">Retro handheld maker Anbernic</a> is <a target="_blank" class="link" href="https://x.com/anbernic_china/status/2014674553310769376" data-i13n="cpos:2;pos:1">launching</a> a new controller with a screen. But unlike Nintendo's Wii U controller, there's no second-screen gaming capability here. Instead, the display is designed to "make customization intuitive and effortless." Oh, and the controller also has a heart rate sensor. Because, hey, if you're going weird, I say fully commit!The Anbernic RG G01 has a 2.5-inch HD display that the company describes as an "HD smart screen." The idea is to remap buttons, program macros and change other settings without requiring software on a connected device.<figure><img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/gamepad_2.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/gamepad_2.jpg" style="height:1440px;width:2560px;" alt="A person's hands holding a gamepad that has a screen on it. It shows their heart rate." data-uuid="013cbf2f-f769-4e75-82fe-d17901116611"><figcaption>You, too, can monitor your heart rate while gaming.</figcaption><div class="photo-credit">Anbernic</div></figure>Meanwhile, Anbernic says the heart rate detection helps you "monitor your well-being during intense sessions." I would hope that folks with heart conditions would use something other than a gamepad to monitor their pulse. And I don’t know why anyone else would be worried about that while gaming. (Too many Red Bulls?) Regardless, Anbernic seems to be leaning into the novelty / curiosity space here, so at least it fits the motif.The RG G01 connects in three ways: Bluetooth 5.0, 2.4Ghz wireless and over a wire. It supports onboard calibration for the triggers, joystick and 6-axis gyroscope. There are four programmable buttons (including macro support) on the backside. The company promises a 1,000Hz polling rate in wired and wireless modes. The gamepad is compatible with PC, Switch, Android and iOS.We don't yet know when this glorious oddity will arrive, other than "coming soon." We also don't know how much it will cost. But you can watch the launch video below and see if it's your type of strange.<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/EisA2TLhUeg?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>This article originally appeared on Engadget at https://www.engadget.com/gaming/retro-handheld-maker-anbernic-has-a-new-gamepad-with-a-screen-and-heart-rate-sensor-201651899.html?src=rss

Jan 23, 2026

Apple will begin showing more App Store ads starting in March

More ads are coming to App Store search results starting in March, Apple shared on <a target="_blank" class="link" href="https://ads.apple.com/app-store/help/ad-placements/0082-search-results" data-i13n="cpos:1;pos:1">an advertising help page</a>. The company <a target="_blank" class="link" href="https://www.engadget.com/apps/apple-is-bringing-even-more-ads-to-the-app-store-193416921.html" data-i13n="cpos:2;pos:1">first said</a> that it would increase the number of App Store ads last month, and this new rollout of search ads will begin on Tuesday, March 3, according to a developer email <a target="_blank" class="link" href="https://www.macrumors.com/2026/01/23/more-app-store-ads-coming-soon/" data-i13n="cpos:3;pos:1">viewed by MacRumors</a>."Search is the way most people find and download apps on the App Store, with nearly 65 percent of downloads happening directly after a search," Apple says. "To help give advertisers more opportunities to drive downloads from search results, Apple Ads will introduce additional ads across search queries." Up until this point, ads for related apps have appeared at the top of search results, but now they'll also appear "further down in search results," according to Apple.App Store activity makes up a significant portion of what Apple calls its <a target="_blank" class="no-affiliate-link link" href="https://www.apple.com/newsroom/2026/01/2025-marked-a-record-breaking-year-for-apple-services/" data-i13n="elm:context_link;elmt:doNotAffiliate;cpos:4;pos:1">"services" business</a>. The company makes money on every App Store transaction, whether it's an app download or an in-app purchase, and increasingly, by selling ad space to companies looking to reach users. App Store ads are hardly new, but the number of ads has steadily increased over the years. Apple <a target="_blank" class="link" href="https://www.engadget.com/apple-app-store-today-tab-ads-152249659.html" data-i13n="cpos:5;pos:1">added ads to the Today tab</a> in 2022 — a space that's already home to editorial curation that doubles as marketing — and in 2025, <a target="_blank" class="link" href="https://www.bloomberg.com/news/newsletters/2025-10-26/what-s-next-for-the-ipad-pro-iphone-17-pro-like-vapor-chamber-apple-maps-ads-mh7nq39h" data-i13n="cpos:6;pos:1">Bloomberg reported</a> the company planned to bring ads to Apple Maps.Apple's decision to rebrand its advertising business from Apple Search Ads to <a target="_blank" class="link" href="https://9to5mac.com/2025/04/14/apple-rebrands-search-ads-business-as-apple-ads/" data-i13n="cpos:7;pos:1">Apple Ads</a> in April 2025 was maybe the best indication that the company was interested in expanding the number of places it would help partners try and reach customers. And it makes sense: the company's billions of devices, each pre-installed with default apps, are some of the most valuable real estate it owns.This article originally appeared on Engadget at https://www.engadget.com/apps/apple-will-begin-showing-more-app-store-ads-starting-in-march-192031226.html?src=rss

Jan 23, 2026

TikTok finalizes deal for its US entity

After years of uncertainty over TikTok's future in the United States, a deal for the app's US business has <a target="_blank" class="link" href="https://newsroom.tiktok.com/announcement-from-the-new-tiktok-usds-joint-venture-llc?lang=en" data-i13n="cpos:1;pos:1">been finalized</a>. The new US entity is called <a target="_blank" class="link" href="https://usdsjv.tiktok.com/" data-i13n="cpos:2;pos:1">TikTok USDS Joint Venture</a>. ByteDance has retained a 19.9 percent percent stake in the new business, with the rest controlled by a group of non-Chinese investors, including Oracle, Silver Lake and MGX, an Emirati-state owned investment firm, all of which have a 15 percent stake. Dell CEO Michael Dell and other investors have smaller stakes in the new company. The terms of the deal were first <a target="_blank" class="link" href="https://www.engadget.com/social-media/trumps-tiktok-deal-is-another-step-closer-to-finally-actually-happening-001813404.html" data-i13n="slk:leaked;cpos:3;pos:1">leaked</a> last month, after TikTok CEO Shou Chew reportedly told employees in a memo that TikTok and ByteDance had agreed to a group of investors. This ends a lengthy saga and months of <a target="_blank" class="link" href="https://www.engadget.com/social-media/trumps-tiktok-deal-is-another-step-closer-to-finally-actually-happening-001813404.html" data-i13n="slk:slow progress;cpos:4;pos:1">slow progress</a> as the agreement was being worked out, ensuring that the app will remain available in the US after years of being on the verge of a ban in the country.President Donald Trump, who had tried to ban the app during his first term in office, praised the deal in a post on Truth Social. "It will now be owned by a group of Great American Patriots and Investors, the Biggest in the World, and will be an important Voice," he <a target="_blank" class="link" href="https://truthsocial.com/@realDonaldTrump/posts/115942147803684675" data-i13n="cpos:5;pos:1">wrote</a>."I only hope that long into the future I will be remembered by those who use and love TikTok."According to TikTok’s announcement, the joint venture will protect American users’ data with Oracle's secure US cloud environment. It will also retrain TikTok’s algorithm on US users’ data and will be in charge of content moderation in the US. The entity promises interoperability, as well, promising that users will still get international content and, if they’re a creator, viewers. “The safeguards provided by the Joint Venture will also cover CapCut, and Lemon8 and a portfolio of other apps and websites in the US,” TikTok said.The new entity will be overseen by a seven-member board of directors, most of whom are Americans. It includes, Shou Chew, the Chief Executive Officer of TikTok, Silver Lake co-CEO Egon Durban, Oracle Executive Vice President Kenneth Glueck and MGX Chief Strategy and Safety Officer David Scott. Adam Presser, who had previously been head of operations and trust and safety at TikTok, is the CEO of TikTok USDS Joint Venture.Exactly what the new joint venture means for US users of TikTok is unclear. Shortly after the deal was announced, TikTok introduced new <a target="_blank" class="link" href="https://www.tiktok.com/legal/page/us/terms-of-service/en" data-i13n="cpos:6;pos:1">terms of service</a> for US users. As the BBC <a target="_blank" class="link" href="https://www.bbc.com/news/articles/cq5yynydvgzo" data-i13n="cpos:7;pos:1">notes</a>, the new terms include provisions relating to use of the app by kids under 13 (they are limited to the "Under 13 Experience") and that the "TikTok USDS Joint Venture does not endorse any content" in the app. The company hasn’t announced specific changes to the app’s algorithm or other core features.Update, January 23, 2026, 10:58AM PT: This post was updated to add a statement from President Trump, and with additional information about TikTok’s new terms of service.This article originally appeared on Engadget at https://www.engadget.com/big-tech/tiktok-finalizes-deal-for-its-us-entity-010543484.html?src=rss

Jan 23, 2026

Vimeo lays off most of its staff just months after being bought by private equity firm

Vimeo just got hit by a brutal round of layoffs, according to a <a data-i13n="cpos:1;pos:1" href="https://www.businessinsider.com/vimeo-laying-off-staff-after-billion-sale-to-bending-spoons-2026-1"><ins>report by </ins><ins>Business Insider</ins></a>. Staffers are <a data-i13n="elm:affiliate_link;sellerN:LinkedIn;elmt:;cpos:2;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=6deefbf2-941b-4156-9f41-a61ebb50d13d&siteId=us-engadget&pageId=1p-autolink&contentUuid=141850da-dc1d-4c95-b098-d03c49952bba&featureId=text-link&merchantName=LinkedIn&linkText=posting&custData=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&signature=AQAAAWwFvX4xIV1l7M6spbEfs1a6ka89bFcfMrCjS4h8E9AN&gcReferrer=https%3A%2F%2Fwww.linkedin.com%2Fposts%2Fsteve-dixon-64928b56_i-was-laid-off-from-vimeo-today-along-with-activity-7419431951738048512-ax3r%2F" class="rapid-with-clickid" data-original-link="https://www.linkedin.com/posts/steve-dixon-64928b56_i-was-laid-off-from-vimeo-today-along-with-activity-7419431951738048512-ax3r/"><ins>posting</ins></a> on <a data-i13n="cpos:3;pos:1" href="https://x.com/daemon404/status/2013988239829303624?s=20"><ins>various social media sites</ins></a> that the layoffs have impacted <a data-i13n="elm:affiliate_link;sellerN:LinkedIn;elmt:;cpos:4;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=6deefbf2-941b-4156-9f41-a61ebb50d13d&siteId=us-engadget&pageId=1p-autolink&contentUuid=141850da-dc1d-4c95-b098-d03c49952bba&featureId=text-link&merchantName=LinkedIn&linkText=most+of+the+company&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5saW5rZWRpbi5jb20vZmVlZC91cGRhdGUvdXJuOmxpOmFjdGl2aXR5Ojc0MTk4NDc1NjkxMzQwNDMxMzcvIiwiY29udGVudFV1aWQiOiIxNDE4NTBkYS1kYzFkLTRjOTUtYjA5OC1kMDNjNDk5NTJiYmEiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3LmxpbmtlZGluLmNvbS9mZWVkL3VwZGF0ZS91cm46bGk6YWN0aXZpdHk6NzQxOTg0NzU2OTEzNDA0MzEzNy8ifQ&signature=AQAAAbRWCof5uxA6KMMAzM7LyQakFmZH2GRjKJ7cE4R5BNgd&gcReferrer=https%3A%2F%2Fwww.linkedin.com%2Ffeed%2Fupdate%2Furn%3Ali%3Aactivity%3A7419847569134043137%2F" class="rapid-with-clickid" data-original-link="https://www.linkedin.com/feed/update/urn:li:activity:7419847569134043137/"><ins>most of the company</ins></a>, including the entire video team. Vimeo is a video-hosting platform, so that sounds bad. “Yesterday, following Vimeo’s recent acquisition by a private equity firm, I learned that I, along with a large portion of the company, was impacted by layoffs,” wrote the company’s former vice president of Global Brand & Creative, Dave Brown. He is referring to a firm called Bending Spoons that bought Vimeo for $1.38 billion <a data-i13n="cpos:5;pos:1" href="https://techcrunch.com/2025/09/10/vimeo-to-be-acquired-by-bending-spoons-in-1-38b-all-cash-deal/"><ins>in the latter half of 2025</ins></a>. We don't know why parent company Bending Spoons conducted such a massive round of layoffs, but the equity firm is known for purchasing tech companies and aggressively cutting costs via layoffs. It did the same thing to Evernote <a data-i13n="cpos:6;pos:1" href="https://www.engadget.com/evernote-is-relocating-to-europe-after-laying-off-most-of-its-us-workforce-205012133.html"><ins>back in 2023</ins></a> and <a data-i13n="cpos:7;pos:1" href="https://www.reuters.com/technology/bending-spoons-lay-off-75-wetransfer-staff-post-deal-ceo-says-2024-09-08/"><ins>WeTransfer in 2024</ins></a>. Engadget has reached out to Vimeo to inquire about the exact number of employees that were laid off and will update this post when we hear back. "I can confirm that a layoff was announced at Vimeo on January 20, 2026. To respect the privacy of those departing, we cannot provide additional details at this time,” a Bending Spoons spokesperson told Gizmodo <a data-i13n="cpos:8;pos:1" href="https://gizmodo.com/vimeo-lays-off-most-of-its-staff-allegedly-includes-the-entire-video-team-2000713416"><ins>in an emailed statement</ins></a>. "Going forward, Bending Spoons remains committed to growing Vimeo to meet the needs of its diverse user base." <div id="bc5b64472a63492f9dc1d171052d28a0"><blockquote class="twitter-tweet">Reviving this account to say: Almost everyone at Vimeo was laid off yesterday, including the entire video team. If you're looking for talented engineers, there are a few on the market. Sucks to see something I built killed by private equity in a technology company skin suit.— Derek Buitenhuis (@daemon404) <a href="https://twitter.com/daemon404/status/2013988239829303624?ref_src=twsrc%5Etfw">January 21, 2026</a></blockquote> </div> It's good to know the company "remains committed to growing Vimeo" after firing just about everyone that works there. One former employee said on X that it "sucks to see something I built killed by private equity in a technology company skin suit." Vimeo has been around a long time. The platform was founded a full year before YouTube and has positioned itself as being a premium alternative for hosting creative and business-adjacent videos. We have no idea what it'll look like with a minimal staff and no video team.This article originally appeared on Engadget at https://www.engadget.com/big-tech/vimeo-lays-off-most-of-its-staff-just-months-after-being-bought-by-private-equity-firm-184556023.html?src=rss

Jan 23, 2026

Tesla paywalls lane centering on new Model 3 and Model Y purchases

Tesla just objectively decreased the value of the Model 3 and Model Y. On Thursday, the company <a target="_blank" class="link" href="https://electrek.co/2026/01/23/tesla-cuts-standard-autopilot-paywalls-basic-safety-feature-behind-fsd-subscription/" data-i13n="cpos:1;pos:1">said</a> it’s paywalling its lane-centering feature, Autosteer, for new purchases of the two EVs in the US and Canada. This was previously a standard feature. Lane centering is now part of the <a target="_blank" class="link" href="https://www.engadget.com/transportation/evs/tesla-used-deceptive-language-to-market-autopilot-california-judge-rules-035826786.html" data-i13n="cpos:2;pos:1">Full Self-Driving Supervised (FSD) package</a>, which costs $99 per month.Speculating on why Tesla would do this doesn't require much imagination. Remember the pay package <a target="_blank" class="link" href="https://www.engadget.com/tesla-shareholders-approve-elon-musks-1-trillion-compensation-package-233217531.html" data-i13n="cpos:3;pos:1">Tesla shareholders approved for Musk</a> in November? One of his requirements for receiving up to $1 trillion is reaching 10 million active FSD subscriptions.<figure><img src="https://media-mbst-pub-ue1.s3.amazonaws.com/creatr-uploaded-images/2025-01/35048eb0-da98-11ef-b7fe-ce074f9aa83d" data-crop-orig-src="https://media-mbst-pub-ue1.s3.amazonaws.com/creatr-uploaded-images/2025-01/35048eb0-da98-11ef-b7fe-ce074f9aa83d" style="height:2648px;width:3972px;" alt="TOPSHOT - Tesla and SpaceX CEO Elon Musk gestures as he speaks during the inaugural parade inside Capitol One Arena, in Washington, DC, on January 20, 2025. (Photo by ANGELA WEISS / AFP) (Photo by ANGELA WEISS/AFP via Getty Images)" data-uuid="df9b41b2-8ec6-39bc-afd7-89641e85684d"/><figcaption>One of Musk's stipulations for receiving $1 trillion is reaching 10 million active FSD subscriptions. (Photo by ANGELA WEISS/AFP via Getty Images)</figcaption><div class="photo-credit">ANGELA WEISS via Getty Images</div></figure>As for the details, Tesla killed the Basic Autopilot feature that was standard on the Model 3 and Model Y. That package included both Autosteer and Traffic-Aware Cruise Control (TACC). Now, the latter will remain a standard feature, but Autosteer now requires that $99-per-month FSD subscription. The change applies only to new purchases, not existing owners.The move comes ahead of <a target="_blank" class="link" href="https://www.engadget.com/transportation/evs/teslas-full-self-driving-is-switching-to-a-subscription-only-service-120053137.html" data-i13n="cpos:4;pos:1">FSD's switch to a subscription-only service</a>. After February 14, you'll no longer be able to buy the package for a one-time $8,000 upfront fee. If you want FSD (or just lane centering) from now on, your annual cost will be $1,188.Making matters worse, Musk warned that FSD won't remain at $99 monthly forever. "I should also mention that the $99/month for supervised FSD will rise as FSD's capabilities improve," the Tesla CEO <a target="_blank" class="link" href="https://x.com/elonmusk/status/2014539856890716418" data-i13n="cpos:5;pos:1">posted</a> on X. He didn't indicate when that might happen or what the increase would be.This article originally appeared on Engadget at https://www.engadget.com/transportation/evs/tesla-paywalls-lane-centering-on-new-model-3-and-model-y-purchases-184000707.html?src=rss

Jan 23, 2026

Meta is temporarily pulling teens' access from its AI chatbot characters

Meta will no longer allow teens to chat with its AI chatbot characters in their present form. The company <a target="_blank" class="link" href="https://about.fb.com/news/2025/10/teen-ai-safety-approach/" data-i13n="slk:announced Friday;cpos:1;pos:1">announced Friday</a> that it will be "temporarily pausing teens’ access to existing AI characters globally."The pause comes months after Meta had said it was working on chatbot-focused <a target="_blank" class="link" href="https://www.engadget.com/ai/meta-is-adding-ai-chatbot-focused-parental-controls-to-instagram-100027229.html" data-i13n="slk:parental controls;cpos:2;pos:1">parental controls</a> following reports that some of Meta's character chatbots had engaged <a target="_blank" class="link" href="https://www.engadget.com/ai/metas-ai-chatbots-were-reportedly-able-to-engage-in-sexual-conversations-with-minors-193726679.html" data-i13n="slk:in sexual conversations;cpos:3;pos:1">in sexual conversations</a> and other alarming interactions with teens. Reuters reported on an internal <a target="_blank" class="link" href="https://www.engadget.com/ai/an-internal-meta-ai-document-said-chatbots-could-have-sensual-conversations-with-children-191101296.html?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAAK8jOG6ZjW9Df42BRFk1UksqV40HVE8Q0wTbOH5IZuP59ZiagOZy1iL3UxuJIEu2MfuOi8q-jZs_dNIKmHAs4qdPGtWziwt_PzQMluCqRQfu_ogi2IkikiavMMFYoH6ITEl6V-WNJ7QEcT1zAs49Sv0IJakYQTh3G8DM-5A36fuc" data-i13n="slk:Meta policy document;cpos:4;pos:1">Meta policy document</a> that said the chatbots were permitted to have "sensual" conversations with underage users, language Meta later said was "erroneous and inconsistent with our policies." The company announced in August that it was <a target="_blank" class="link" href="https://www.engadget.com/social-media/meta-is-re-training-its-ai-so-it-wont-discuss-self-harm-or-have-romantic-conversations-with-teens-182418587.html" data-i13n="slk:re-training;cpos:5;pos:1">re-training</a> its character chatbots to add "guardrails as an extra precaution" that would prevent teens from discussing self harm, disordered eating and suicide. Now, Meta says it will prevent teens from accessing any of its character chatbots until "the updated experience is ready." Those updates will include parental controls, according to a Meta spokesperson. The new restrictions, which will be starting "in the coming weeks," will apply to those with teen accounts, "as well as people who claim to be adults but who we suspect are teens based on our age prediction technology." Teens will still be able to access the official Meta AI chatbot, which the company says already has "age-appropriate protections in place." Meta and other AI companies that make "companion" characters have faced increasing scrutiny over the safety risks these chatbots could pose to young people. The <a target="_blank" class="link" href="https://www.engadget.com/ai/the-ftc-is-investigating-companies-that-make-ai-companion-chatbots-181413615.html" data-i13n="slk:FTC;cpos:6;pos:1">FTC</a> and the Texas <a target="_blank" class="link" href="http://engadget.com/ai/texas-ag-to-investigate-meta-and-characterai-over-misleading-mental-health-claims-221343275.html" data-i13n="slk:attorney general;cpos:7;pos:1">attorney general</a> have both kicked off investigations into Meta and other companies in recent months. The issue of chatbots has also come up in the context of a <a target="_blank" class="link" href="https://www.engadget.com/facebook-and-instagrams-algorithms-facilitated-child-sexual-harassment-state-lawsuit-claims-095314139.html" data-i13n="slk:safety lawsuit;cpos:8;pos:1">safety lawsuit</a> brought by New Mexico's attorney general. A trial is scheduled to start early next month; Meta's lawyers have attempted to exclude testimony related to the company's AI chatbots, Wired <a target="_blank" class="no-affiliate-link link" href="https://www.wired.com/story/meta-child-safety-trial-ask-judge-bar-mental-health-harvard/" data-i13n="elm:context_link;elmt:doNotAffiliate;slk:reported;cpos:9;pos:1">reported</a> this week.Correction, January 23, 2026, 11:18AM PT: This post was updated to clarify that Meta’s planned chatbot parental control features have not yet rolled out. This article originally appeared on Engadget at https://www.engadget.com/social-media/meta-is-temporarily-pulling-teens-access-from-its-ai-chatbot-characters-180626052.html?src=rss

Jan 23, 2026

You can now create AI-generated coloring books in Microsoft Paint

Microsoft CEO Satya Nadella recently went on <a data-i13n="cpos:1;pos:1" href="https://www.windowscentral.com/artificial-intelligence/microsoft-ceo-satya-nadella-says-ai-needs-to-prove-its-worth"><ins>record</ins></a> saying that AI still needs to prove its worth if society is to adopt it long-term, but he presumably thinks his company has cracked it with its latest innovation: AI coloring books. A new Microsoft Paint <a data-i13n="cpos:2;pos:1" href="https://blogs.windows.com/windows-insider/2026/01/21/notepad-and-paint-updates-begin-rolling-out-to-windows-insiders/">feature</a> currently rolling out to Windows Insiders allows you to generate coloring book pages based on the text prompt you enter. The example Microsoft uses is "a cute fluffy cat on a donut," to which the AI tool will spit out a set of slightly different options based on your prompt. You can then choose which image you want, add it to your current workspace, copy or save it. Presumably you can also print it out for the purpose of entertaining your kids. No doubt the kind of real-world impact the Microsoft chief was alluding to. The coloring book feature is exclusive to Copilot+ PCs, and Microsoft is also adding a fill tolerance slider that lets you adjust the precision with which the Fill tool adds color to your canvas. As well as Paint’s new Coloring book feature, Microsoft has also improved its Write, Rewrite and Summarize AI functionality in <a data-i13n="elm:affiliate_link;sellerN:;elmt:;cpos:3;pos:1" href="https://shopping.yahoo.com/rdlw?siteId=us-engadget&pageId=1p-autolink&contentUuid=9c684b1e-34b7-44ba-84cc-28ec2fd24327&featureId=text-link&linkText=Notepad&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3N1cHBvcnQubWljcm9zb2Z0LmNvbS9lbi1nYi93aW5kb3dzL2VuaGFuY2UteW91ci13cml0aW5nLXdpdGgtYWktaW4tbm90ZXBhZC00MDg4Yjk1NC1jOTdiLTQ2ZGMtODEzZi05NTliZTAxNzQ2ZDUiLCJjb250ZW50VXVpZCI6IjljNjg0YjFlLTM0YjctNDRiYS04NGNjLTI4ZWMyZmQyNDMyNyIsIm9yaWdpbmFsVXJsIjoiaHR0cHM6Ly9zdXBwb3J0Lm1pY3Jvc29mdC5jb20vZW4tZ2Ivd2luZG93cy9lbmhhbmNlLXlvdXItd3JpdGluZy13aXRoLWFpLWluLW5vdGVwYWQtNDA4OGI5NTQtYzk3Yi00NmRjLTgxM2YtOTU5YmUwMTc0NmQ1In0&signature=AQAAAQEZpndR7_4OdS5MKacGLTXxdX9Xd4AE5PFU1QmURLAu&gcReferrer=https%3A%2F%2Fsupport.microsoft.com%2Fen-gb%2Fwindows%2Fenhance-your-writing-with-ai-in-notepad-4088b954-c97b-46dc-813f-959be01746d5" class="rapid-with-clickid" data-original-link="https://support.microsoft.com/en-gb/windows/enhance-your-writing-with-ai-in-notepad-4088b954-c97b-46dc-813f-959be01746d5"><ins>Notepad</ins></a>, which integrates with GPT to fine-tune your writing and summarize complex notes. You’ll need to sign into your Microsoft account to use cloud features, but results will now appear more quickly and let you interact with the preview without having to wait for its full response. Again, you’ll need to be Windows Insider in the Canary and Dev channels on Windows 11 to take advantage of the updates initially.This article originally appeared on Engadget at https://www.engadget.com/ai/you-can-now-create-ai-generated-coloring-books-in-microsoft-paint-163512527.html?src=rss

Jan 23, 2026

NBA League Pass is up to 55 percent off

Basketball fans can save on <a data-i13n="elm:affiliate_link;sellerN:National Basketball Association;elmt:;cpos:1;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=c109eac4-825f-483e-8edf-51532ca9e7cb&siteId=us-engadget&pageId=1p-autolink&contentUuid=a364f8fe-4cef-4bc2-89cf-730e7b34af39&featureId=text-link&merchantName=National+Basketball+Association&linkText=NBA+League+Pass&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5uYmEuY29tL2xlYWd1ZS1wYXNzLXB1cmNoYXNlIiwiY29udGVudFV1aWQiOiJhMzY0ZjhmZS00Y2VmLTRiYzItODljZi03MzBlN2IzNGFmMzkiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3Lm5iYS5jb20vbGVhZ3VlLXBhc3MtcHVyY2hhc2UifQ&signature=AQAAAdUhKaPYHJzinvfA7MLKAWA2Ghk-P2COLZr_GV2bgBXf&gcReferrer=https%3A%2F%2Fwww.nba.com%2Fleague-pass-purchase" class="rapid-with-clickid" data-original-link="https://www.nba.com/league-pass-purchase">NBA League Pass</a> right now, which lets you catch a bunch of out-of-market NBA games via streaming. The League Pass Premium subscription is on sale for $75, down from the usual $160, and League Pass Standard is marked down to $50 from $110. Considering we're almost halfway though the season, the discount makes sense and is a good deal for anyone who wants to keep a close eye on the rest of the games to be played this year. The Standard plan includes commercials and support for only one device at a time, while the Premium tier offers no commercials, in-arena streams during breaks in the game, offline viewing of full games and concurrent streams on up to three devices at once. <core-commerce id="5e40023bdb824a11adc46b45eac49663" data-type="product-list" data-original-url="https://www.nba.com/league-pass-purchase"></core-commerce> Last year, League Pass <a data-i13n="cpos:2;pos:1" href="https://www.engadget.com/entertainment/nba-league-pass-gets-multiview-just-in-time-for-the-season-to-start-191016755.html">added multiview</a>, which allows you to view up to four games at once on a single screen. This is included across both subscription tiers. The service also added a smart rewind tool that automatically selects key highlights and plays from each game. Outside the US and Canada, League Pass carries every single NBA game live, but within these countries a bevy of restrictions apply. In the US, any games being shown on your regional sports network will be blacked out as the service is meant for out-of-market games only. Also, any nationally broadcast games will not be available live, but instead will be available for on-demand viewing at 6AM ET the following day. The service is only for regular-season games. If you're an avid NBA fan that follows multiple teams then the League Pass almost certainly carries dozens of games you can watch even with the restrictions in the US. Subscribers can get a list of applicable blackouts by entering their ZIP code before signing up. Follow <a data-i13n="cpos:3;pos:1" href="https://twitter.com/EngadgetDeals">@EngadgetDeals</a> on X for the latest <a data-i13n="cpos:4;pos:1" href="https://www.engadget.com/deals/">tech deals</a> and <a data-i13n="cpos:5;pos:1" href="https://www.engadget.com/best-tech/">buying advice</a>.This article originally appeared on Engadget at https://www.engadget.com/deals/nba-league-pass-is-up-to-55-percent-off-163421368.html?src=rss

Jan 23, 2026

Malaysia lifts ban on Grok after taking X at its word

After being one of the first countries in the world to block Elon Musk’s Grok chatbot, Malaysia has now lifted its ban. Along with Indonesia, the country moved swiftly to temporarily <a data-i13n="cpos:1;pos:1" href="https://www.theguardian.com/technology/2026/jan/12/malaysia-blocks-elon-musk-grok-ai-fake-sexualised-images-indonesia-x-chatbot"><ins>halt access</ins></a> to X's frequently controversial AI chatbot earlier this month, after multiple <a data-i13n="cpos:2;pos:1" href="https://www.engadget.com/ai/elon-musks-grok-ai-posted-csam-image-following-safeguard-lapses-140521454.html"><ins>reports</ins></a> emerged of it being used to generate deepfake sexualized images of people, including women and children. At the time, the Malaysian Communications and Multimedia Commission (MCMC) said the restrictions would remain in place until X Corp and parent xAI could prove they had enforced the necessary safeguards against misuse of the above nature. Malaysian authorities appear to be taking X at its word, after the MCMC released a <a data-i13n="cpos:3;pos:1" href="https://x.com/MCMC_RASMI/status/2014568371405390194?s=20"><ins>statement</ins></a> confirming it was satisfied that Musk’s company has implemented the required safety measures. It added that the authorities will continue to monitor the social media platform, and that any further user safety breaches or violations of Malaysian laws would be dealt with firmly. At the time of writing, only Malaysia and Indonesia have hit Grok with official bans, though UK regulator Ofcom opened a formal <a data-i13n="cpos:4;pos:1" href="https://www.engadget.com/big-tech/uk-regulator-ofcom-opens-a-formal-investigation-into-x-over-csam-scandal-120000312.html"><ins>investigation</ins></a> into X under the country’s Online Safety Act, in the wake of the non-consensual sexual deepfake scandal. X has since <a data-i13n="cpos:5;pos:1" href="https://www.engadget.com/ai/x-says-grok-will-no-longer-edit-images-of-real-people-into-bikinis-231430257.html"><ins>changed</ins></a> its image-editing policies, and on January 14 the company said Grok will no longer allow "the editing of images of real people in revealing clothing such as bikinis." Earlier this week, the UK-based non-profit, the Center for Countering Digital Hate (CCDH), <a data-i13n="cpos:6;pos:1" href="https://counterhate.com/research/grok-floods-x-with-sexualized-images/"><ins>estimated</ins></a> that in the 11-day period between December 29 and January 9, <a data-i13n="cpos:7;pos:1" href="https://www.engadget.com/ai/grok-generated-an-estimated-3-million-sexualized-images--including-23000-of-children--over-11-days-175053250.html"><ins>Grok</ins></a> generated approximately 3 million sexualized images, around 23,000 of which were of children.This article originally appeared on Engadget at https://www.engadget.com/ai/malaysia-lifts-ban-on-grok-after-taking-x-at-its-word-144457468.html?src=rss

Jan 23, 2026

Get the 512GB Samsung P9 microSD Express card for 33 percent off right now

MicroSD Express cards are still a little hard to find, considering they're pretty new and only really started becoming popular last year once the Switch 2 came out. These upgraded versions of microSD cards are the only ones compatible with the Switch 2 for expanding its storage, so if you're already starting to feel the crunch on your console, it's worth picking one up. <a data-i13n="elm:affiliate_link;sellerN:Amazon;elmt:;cpos:1;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=66ea567a-c987-4c2e-a2ff-02904efde6ea&itemId=amazon_B0FT995SPF&siteId=us-engadget&pageId=1p-autolink&contentUuid=b7c1c9b1-8762-4b3b-bd34-558e13eba66f&featureId=text-link&merchantName=Amazon&linkText=Samsung%27s+P9+microSD+Express+card&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5hbWF6b24uY29tL1NhbXN1bmctRXhwcmVzcy1taWNyb1NEWEMtTmludGVuZG8tU3dpdGNoLUFNL2RwL0IwRlQ5OTVTUEY_dGFnPWdkZ3QwYy0yMCIsImNvbnRlbnRVdWlkIjoiYjdjMWM5YjEtODc2Mi00YjNiLWJkMzQtNTU4ZTEzZWJhNjZmIiwib3JpZ2luYWxVcmwiOiJodHRwczovL3d3dy5hbWF6b24uY29tL1NhbXN1bmctRXhwcmVzcy1taWNyb1NEWEMtTmludGVuZG8tU3dpdGNoLUFNL2RwL0IwRlQ5OTVTUEYiLCJkeW5hbWljQ2VudHJhbFRyYWNraW5nSWQiOnRydWUsInNpdGVJZCI6InVzLWVuZ2FkZ2V0IiwicGFnZUlkIjoiMXAtYXV0b2xpbmsiLCJmZWF0dXJlSWQiOiJ0ZXh0LWxpbmsifQ&signature=AQAAAT_twTWcajbKqdvcyEBHoh5RgsQUOS3jFfNNm7DTjYub&gcReferrer=https%3A%2F%2Fwww.amazon.com%2FSamsung-Express-microSDXC-Nintendo-Switch-AM%2Fdp%2FB0FT995SPF" class="rapid-with-clickid" data-original-link="https://www.amazon.com/Samsung-Express-microSDXC-Nintendo-Switch-AM/dp/B0FT995SPF?th=1">Samsung's P9 microSD Express card</a> is on sale right now — you can grab the 512GB version of $80, which is 33 percent off and one of the best prices we've seen. <core-commerce id="91bfde4ef7fd4c03825d95a9012acdc5" data-type="product-list" data-original-url="https://www.amazon.com/Samsung-Express-microSDXC-Nintendo-Switch-AM/dp/B0FT995SPF?th=1"></core-commerce> The P9 boasts transfer speeds of up to 800MB/s, making moving games to the card that much faster. As for load times, in our testing we found that any microSD Express, the standard the <a data-i13n="cpos:2;pos:1" href="https://www.engadget.com/gaming/nintendo/switch-2/">Switch 2</a> requires, will offer roughly the same performance. This format is pretty new, so there aren't a ton of cards on the market. As such, the P9 makes our list of <a data-i13n="cpos:3;pos:1" href="https://www.engadget.com/gaming/nintendo/best-microsd-cards-for-nintendo-switch-2-160052947.html">best microSD cards</a> for the Nintendo Switch 2. The P9 microSD Express is also compatible with the Steam Deck or any other gaming console that accepts the format, as well as cameras and more. Follow <a data-i13n="cpos:4;pos:1" href="https://twitter.com/EngadgetDeals">@EngadgetDeals</a> on X for the latest <a data-i13n="cpos:5;pos:1" href="https://www.engadget.com/deals/">tech deals</a> and <a data-i13n="cpos:6;pos:1" href="https://www.engadget.com/best-tech/">buying advice</a>.This article originally appeared on Engadget at https://www.engadget.com/deals/get-the-512gb-samsung-p9-microsd-express-card-for-33-percent-off-right-now-143849920.html?src=rss

Jan 23, 2026

Samsung Bespoke Fridge with AI review: All the bells and whistles

At their core, refrigerators are relatively simple devices. If you're the type of person to view every extra feature as a component that could potentially go wrong, basic iceboxes are probably the kind you go for. But for those on the other end of the spectrum, Samsung's latest Bespoke Refrigerators with AI inside have more bells and whistles than you might think possible — including an optional 32-inch screen. Editor's note: The model we tested for this review came out in the second half of 2025 and will continue to be on sale throughout 2026. Hardware will remain the same, the only changes will come in the form of an <a target="_blank" class="link" href="https://news.samsung.com/us/samsung-unveil-ai-vision-google-gemini-ces-2026/#_ftn2" data-i13n="cpos:1;pos:1">OTA software update</a> slated for later this year that will add support for Google Gemini, improved food recognition/labeling and more.  <core-commerce data-type="product-list" id="362e34cb-37bc-447d-a964-0a6c4f1d9b00" data-original-url="https://www.samsung.com/us/home-appliances/refrigerators/bespoke/bespoke-23-cu-ft-counter-depth-4-door-french-door-refrigerator-with-ai-family-hub-ai-vision-inside-in-white-glass-rf90f23aewaa/"></core-commerce> <h2 id="jump-link-design-as-sleek-as-a-fridge-can-be">Design: As sleek as a fridge can be</h2> If you're into faux vintage appliances like <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?siteId=us-engadget&pageId=1p-autolink&contentUuid=1511d80f-6cd2-4573-bcf8-da54b7c99e56&featureId=text-link&linkText=Smeg%27s+retro-inspired+models&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5zbWVnLmNvbS91cy9yZWZyaWdlcmF0b3JzL3JldHJvLXN0eWxlIiwiY29udGVudFV1aWQiOiIxNTExZDgwZi02Y2QyLTQ1NzMtYmNmOC1kYTU0YjdjOTllNTYiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3LnNtZWcuY29tL3VzL3JlZnJpZ2VyYXRvcnMvcmV0cm8tc3R5bGUifQ&signature=AQAAAavr7TPFdnTg0rml4MkUeR_iWqpfFYs3raAa6l5_DRHT&gcReferrer=https%3A%2F%2Fwww.smeg.com%2Fus%2Frefrigerators%2Fretro-style" data-i13n="elm:affiliate_link;sellerN:;elmt:;cpos:2;pos:1" data-original-link="https://www.smeg.com/us/refrigerators/retro-style">Smeg's retro-inspired models</a>, Samsung's offerings might not be for you. For everyone else, the Bespoke family does minimalist design better than practically any other line of fridges on the market. Granted, the counter-depth <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=f85e63cd-e13c-4f9d-991c-9fbaadede3ac&siteId=us-engadget&pageId=1p-autolink&contentUuid=1511d80f-6cd2-4573-bcf8-da54b7c99e56&featureId=text-link&merchantName=Samsung+Electronics&linkText=French+Door+AI+model&custData=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&signature=AQAAATBqsWyCme8yYU5Segus2oK1Il9EJQXvAeuQTOllPemX&gcReferrer=https%3A%2F%2Fwww.samsung.com%2Fus%2Fhome-appliances%2Frefrigerators%2Fbespoke%2Fbespoke-23-cu-ft-counter-depth-4-door-french-door-refrigerator-with-ai-family-hub-ai-vision-inside-in-charcoal-glass-and-stainless-steel-rf90f23aecraa%2F" data-i13n="elm:affiliate_link;sellerN:Samsung Electronics;elmt:;cpos:3;pos:1" data-original-link="https://www.samsung.com/us/home-appliances/refrigerators/bespoke/bespoke-23-cu-ft-counter-depth-4-door-french-door-refrigerator-with-ai-family-hub-ai-vision-inside-in-charcoal-glass-and-stainless-steel-rf90f23aecraa/">French Door AI model</a> I tested is a bit of a departure thanks to a huge 32-inch screen embedded in its right-side door. That said, there are a ton of configurations, including the choice of a smaller 9-inch panel or models without a display at all. The latter looks like a big rectangle, but in a way that Dieter Rams would probably appreciate (I'm just guessing, I don't have his contact info). Other options include setups ranging from four-door models to side-by-side two-door variants or even a single-panel upright model.  <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-front.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-front.jpg" style="height:1200px;width:1920px;" alt="Glare is one of the tiny downsides of a fridge with a glass front. " data-uuid="f1d4f78c-6f26-4f88-9c24-ef3b7177f456"> <figcaption> Glare is one of the tiny downsides of a fridge with a glass front. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> On my review unit, the fresh food section is a touch smaller (16.5 cubic feet) than you might expect from a fridge this size, though I've never really felt limited by the amount of available storage. Samsung does a good job of optimizing interior space by providing a number of adjustable shelves, including a sliding one on the right side that makes it easy to store taller items without completely sacrificing flat space for others. The shelves are also spill proof and made out of tempered glass, so they resist scratches and just look a bit nicer than those made from lower-quality materials. I also appreciate how Samsung found a spot between the two crisper drawers for the water filter, instead of just sticking it in the top corner like a lot of competitors.  <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-inside.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-inside.jpg" style="height:1200px;width:1920px;" alt="Even with the FlexZone drawer stealing some space, the main section of this Bespoke model still features 16.5 cubic feet of storage. " data-uuid="3b973ffe-277f-47c0-9e41-b858d48b5b80"> <figcaption> Even with the FlexZone drawer stealing some space, the main section of this Bespoke model still features 16.5 cubic feet of storage. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> Meanwhile, in exchange for a slightly smaller main cabinet, this model has what Samsung calls a FlexZone drawer, which might be my second favorite thing about this fridge. It's perfectly designed for storing things like 12-ounce cans and juice boxes, and as someone who always likes to keep a couple of chilled Coke Zeros on standby, it's super convenient. That said, taller bottled drinks may need to be stored on their sides, which doesn't feel quite as efficient. But really, you can put anything you want in there and if you have young children like me, it's an excellent place to keep refrigerated kids' snacks. My four-year-old can go in there on his own without needing help from an adult to reach stuff higher up. Samsung even gives you the ability to set a different temperature for the FlexDrawer, which is a nice touch. <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-flexzone.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-flexzone.jpg" style="height:1200px;width:1920px;" alt="The optional FlexZone drawer is amazing for stashing drinks or having a refrigerated section that's easy for kids to reach. " data-uuid="9bd12016-1ec5-4c49-bc9a-3bee4905b454"> <figcaption> The optional FlexZone drawer is amazing for stashing drinks or having a refrigerated section that's easy for kids to reach. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> Additionally, along with some rivals from LG, this fridge also has touch sensors at the bottom of each door that let you open them with a light tap. While I still prefer to just grab the built-in handles (which are located both above and below each door), this is a nice inclusion, particularly on a fridge that doesn't have giant handgrips bolted to the outside. My one small gripe with Samsung's exterior design is that its top isn't completely flat as it dips down in back. Admittedly, this hasn't stopped me from storing stuff like sheet pans and roasting trays above the fridge, but it would be nicer if things were fully level up there.  <h2 id="jump-link-water-and-ice-dispensing-samsungs-beverage-center-reigns-supreme">Water and ice dispensing: Samsung's Beverage Center reigns supreme</h2> I've already gone <a target="_blank" class="link" href="https://www.engadget.com/home/smart-home/samsungs-beverage-center-is-the-best-fridge-feature-competitors-cant-copy-130000113.html" data-i13n="cpos:4;pos:1">on record about how Samsung's Beverage Center</a> is the best new feature to land on any fridge in years, so without belaboring the point, but I'll happily wax poetic about it again. Instead of having a water dispenser installed smack dab on the outside of the door, Samsung opted to hide things behind an exterior flap on the left side of the fridge. This accomplishes several things. Not only does it feel a bit more hygienic because the spigot isn't subjected directly to all the germs swirling around the kitchen or a cough or sneeze from anyone passing by, it also simplifies the fridge's exterior design. This might sound a bit weird, but I've always thought that putting water dispensers on the outside of a fridge was kind of like installing a shower in the middle of a room. It's just kind of awkward. Furthermore, by opening the Beverage Center's door, you also have access to one of the fridge's interior shelves, which makes it quick and easy to grab things like ketchup, mustard or any other sauces you might need for a meal. My wife stashes her matcha there, so it's readily available first thing in the morning.  <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-beverage-center.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-beverage-center.jpg" style="height:1200px;width:1920px;" alt="Samsung's Beverage Center is one of the best fridge innovations and years. It includes both a water dispenser and a built-in pitcher that refills itself automatically. " data-uuid="8bd6a819-f9d3-45ce-afac-742f4d4ebcf0"> <figcaption> Samsung's Beverage Center is one of the best fridge innovations and years. It includes both a water dispenser and a built-in pitcher that refills itself automatically. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> But the most important thing is that right next to the spigot, there's also a built-in water pitcher that automatically refills itself after every use. This means there's pretty much always 40+ ounces of ice-cold water (depending on the specific model) that you can pull out and pour. This allows you to fill up everyone's cup in seconds like a server at a restaurant, instead of blocking the entire fridge while you slowly top off each glass individually. To me, it's the difference between an active process where you can dictate your own pace versus having to stand there while the machine takes its sweet time. Plus, if you like infusing your water with fruit or tea, there's a built-in basket filter for that too. The Beverage Center is a truly useful innovation and one of Samsung’s biggest advantages over its rivals, and because <a target="_blank" class="link" href="https://patents.google.com/patent/US9085453B2/en" data-i13n="cpos:5;pos:1">the company patented it</a>, it'll be years before competitors can come up with copies. <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-ice.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-ice.jpg" style="height:1200px;width:1920px;" alt="The top shelf in the freezer also contains the ice maker, which comes with an included scoop and the ability to make two difference kinds of cubes. " data-uuid="5022dd43-2c59-4715-b748-34a18dc29220"> <figcaption> The top shelf in the freezer also contains the ice maker, which comes with an included scoop and the ability to make two difference kinds of cubes. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> As for ice, the model I'm testing can produce two different types: regular cubes and smaller pebbles. There's also an included scooper and a dedicated shelf for storing it. Now I will admit that the smaller pebbles aren't quite as good as what you get from a standalone nugget-style icemaker, but in my experience they're more than good enough for anyone who likes variety in their frozen water. My small critique for Samsung is that I've heard the ice trays on other models come with sliders that have actual ball bearings instead of plastic rails and I wish the latter were a standard inclusion on every model. There's never so much friction that anything would get properly stuck, it would just be nice if things slid a bit more smoothly. <h2 id="jump-link-the-screen-a-welcome-addition-to-a-digital-lifestyle">The screen: A welcome addition to a digital lifestyle</h2> <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-memos.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-memos.jpg" style="height:1200px;width:1920px;" alt="The fridge's 32-inch display can be used to write memos or draw, along with a ton of other functions. " data-uuid="11dfc002-04dd-4d10-9580-297a87b4fd4b"> <figcaption> The fridge's 32-inch display can be used to write memos or draw, along with a ton of other functions. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> Of course, one of the biggest potential objections about Samsung's latest fridges may also be one of its biggest strengths: their optional displays. With the continued intrusion of screens into our daily lives, I'm not here to tell you that you need one more. However, even though I was skeptical at first, I've grown to really appreciate having one on the fridge and I'd argue that unless you're skipping the feature entirely, the larger 32-inch panel is what you want.  The main reason is that, at least for my family, so much of what we do is already digital. Things like calendars, shopping lists, recipes and family photos already live on my phone and being able to share all of that with the fridge directly via Wi-Fi is a major plus. Frankly, it's easier to upload a picture to the fridge's display than it is to print one out and post it, which goes double for Samsung's Bespoke line since its glass front means magnets don't stick. Even though I already have a smart display in our kitchen, the size of Samsung's screen, along with support for a handful of important apps, have turned the fridge into a better family hub than anything else we own. Because there's so much space, there's room for photos, memos, the weather, filter status and a full 30-day calendar — the latter of which has significantly improved my family's planning and coordination. <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-tv.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-tv.jpg" style="height:1200px;width:1920px;" alt="You can even watch TV or YouTube videos right on the fridge's display. " data-uuid="5d1f2938-d507-43d9-9e9c-cd9e0f0bd59c"> <figcaption> You can even watch TV or YouTube videos right on the fridge's display. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> The fridge runs Samsung's Tizen OS, which is basically the company's lightweight take on Android, so it's super easy to use. You can swipe left and right to add additional widgets to your screen, play music or even watch videos (YouTube, Samsung TV Plus, etc.) if you're so inclined. That said, sometimes I wish the screen was a bit more responsive, because even though I don't think it's a dealbreaker, after having reviewed several generations of the company's phones, I know Samsung can do better. There are also some settings from its handsets that I wish would get ported over to its fridges — like a dedicated dark mode. When I grab a glass of water later at night, the last thing I need is the entire display blasting my kitchen like a pair of hi-beams. <h2 id="jump-link-ai-inside-a-work-in-progress-with-benefits-today">AI Inside: A work in progress with benefits today</h2> With seemingly every tech company trying to shoehorn AI into its latest products, it's easy to roll your eyes at the idea of adding that into a fridge as well. And I think even Samsung would admit that its efforts to integrate machine learning into the Bespoke line are a work in progress. I've found that while it certainly isn't essential, the AI features found here can be a handy bonus in the right situations.  <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-camera.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-camera.jpg" style="height:1200px;width:1920px;" alt="Bespoke models with AI feature a camera mounted to the top of the fridge that can recognize and take pictures of the food you put inside. " data-uuid="ac96cc37-53e3-4d9c-abd7-888be6ca0539"> <figcaption> Bespoke models with AI feature a camera mounted to the top of the fridge that can recognize and take pictures of the food you put inside. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> AI Vision is the headline feature in Samsung's package, which uses machine learning and a camera mounted to the roof of the fridge to automatically recognize and record items you put inside. However, there are some limitations. Currently, Samsung's AI can detect 37 different food items like fresh fruit and veggies, along with an additional 50 kinds of packaged goods like yogurt. Furthermore, the camera only records what goes into the fridge's main cabinet, so anything you put in the FlexZone drawer, freezer or either door won't get recorded. Regardless of whether an item is recognized or not, the fridge will then take a photo of the item (and tag it with the name of the food if it can be properly identified). You can then view a list of everything in the fridge via a dedicated tab on its screen (assuming your model has one) or from your phone via Samsung's SmartThings app. From there, you can do things like set names for anything uncategorized or add an expiration date, though I wish the fridge would do that last part on its own. <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-ai-vision.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-ai-vision.jpg" style="height:1200px;width:1920px;" alt="The fridge uses AI to automatically track food that it can recognize and applies labels to the corresponding photo. " data-uuid="7a1c43f1-40b2-4627-a716-1d2b67ccaace"> <figcaption> The fridge uses AI to automatically track food that it can recognize and applies labels to the corresponding photo. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> In practice, this means you just put items in the fridge like you would normally and let the AI sort things out on its own. You can always go through the main list to see if you have a specific item or let the fridge automatically delete things when you take them out. That said, the AI doesn't get things right every time, because while Samsung's programming does clear things out after a certain amount of time, you will still need to delete food manually from time to time. Thankfully, for items in the door, the fridge can show you a live feed of what's there, so if you're at the store and can't remember if you have milk or not, you can simply pull out your phone and see for yourself.  Because the camera is mounted to the top of the fridge, you might not get the clearest picture or you may end up with your hand covering part of the food as the fridge takes a picture as you put items inside. Even so, that's still more than enough for me to be able to tell what something is just by looking at a small icon. I've found that Samsung AI Vision legitimately makes it easier to grocery shop by allowing me to see what food I have at a glance, regardless of whether I'm at home or at the store.  <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-meal-planning.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-meal-planning.jpg" style="height:1200px;width:1920px;" alt="Samsung also uses AI to suggest recipes based on ingredients you already have in the fridge. You can even specific things like food preferences and dietary restrictions. " data-uuid="49790c31-9501-40cc-966f-ddfa832ed81c"> <figcaption> Samsung also uses AI to suggest recipes based on ingredients you already have in the fridge. You can even specific things like food preferences and dietary restrictions. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> On top of that, the fridge also has built-in meal planning capabilities. This means you can ask the fridge to come up with a dish based on ingredients you already have, and it's surprisingly good. Not only do you have the ability to specify preferences and dietary restrictions, there's a wide variety of recipes it can choose from. On several occasions, I even found that when asked to suggest a recipe, the fridge came up with dishes I already had in mind. Alternatively, it will highlight when you're missing items for a potential dish, so you can grab them the next time you're at the store. This is one of the most helpful features, because often at the end of the week, I find myself with a handful of leftover ingredients. So when the fridge uses that info to suggest recipes that use those items, it helps reduce food waste.  Sadly, the weakest part of Samsung's AI is Bixby. In addition to being overeager when listening for voice prompts, which ended up causing a lot of inadvertent triggers, Samsung also put the Bixby icon near the bottom left corner of the screen. That's right next to the sensor and handle for the door, resulting in even more unwanted summons. After using this fridge for four months, I wish Bixby would just go away, and it seems Samsung agrees. Sometime later this year, the company is adding support for Google Gemini, which should deliver a massive upgrade to the fridge's voice-controlled capabilities. <h2 id="jump-link-a-word-on-ads">A word on ads</h2> <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-ads.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-ads.jpg" style="height:1200px;width:1920px;" alt="You can disable ads that show up on the fridge's cover screen, though that will also get rid of the news and weather widgets. Thankfully that doesn't impact that widgets on the fridge's main home screen. " data-uuid="677abb77-8b22-4d58-8f07-f4f361863e11"> <figcaption> You can disable ads that show up on the fridge's cover screen, though that will also get rid of the news and weather widgets. Thankfully that doesn't impact that widgets on the fridge's main home screen. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> Last fall, to the dismay of many, Samsung announced that it was <a target="_blank" class="link" href="https://www.engadget.com/home/smart-home/samsungs-very-expensive-family-hub-fridges-will-now-treat-you-to-ads-on-their-displays-143529070.html?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAAF28k3xfmfRkWfn8bQMnUSxfDPZY9wL8OsZ7JT-dWUzdrdgmAvzix3Bz3u6VkPZFCE0LP3gq1x8bv-Ois4rQxn9jJdajrBk-pRU-tJh5ICn4rDEGc7J7lXg9sC3_g3vcz420K9rZFclmfNd5kKGq6qWMWRPoJzUOUar40FYE7jxk" data-i13n="cpos:6;pos:1">adding ads to its Family Hub fridges</a>, including Bespoke models with built-in displays. Now no one likes being taunted with commercials, especially when they're on expensive appliances located in the busiest room of their home. Thankfully, I've found that these ads only show up on the Cover Screen, which only appears after no one has used the fridge in a while, kind of like a screensaver on a desktop. Furthermore, you can disable these ads entirely. The downside is that doing so also gets rid of the rotating news and weather widgets that appear on the home screen. But in my experience, that's just fine, because I don't need what is effectively the fridge's lock screen to have that much going on. Plus, when you open the fridge or touch its screen, it instantly reverts to the main home menu, which includes its own set of widgets that you can customize how you like. <h2 id="jump-link-some-quick-notes-and-observations">Some quick notes and observations</h2> <figure> <img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-beverage-center-door.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/bespoke-beverage-center-door.jpg" style="height:1200px;width:1920px;" alt="For some reason, the door to the Beverage Center doesn't soft close like the rest of the doors do. " data-uuid="67cf6a7a-4612-483d-a892-5915f928ab98"> <figcaption> For some reason, the door to the Beverage Center doesn't soft close like the rest of the doors do. </figcaption> <div class="photo-credit"> Sam Rutherford for Engadget </div> </figure> The Bespoke fridge’s screen does collect smudges, but so does stainless steel, and it doesn't really impact viewability. And on white glass models like my review unit, the simple design makes it easier to clean and maintain than typical metal-clad rivals.  The door to the Beverage Center doesn't soft-close like you'd normally expect on a fridge. It's more of a clack, which isn't a bad thing on its own, though I hope Samsung changes its design to match the other doors on future models for consistency's sake.  Because the fridge runs Tizen and not full Android, there aren't nearly as many choices for widgets. For example, there's a widget for Spotify and Amazon Music, but not YouTube or Apple Music. There's also only a single option for weather and a widget for pinning your favorite websites, but very little choice when it comes to dedicated news options. In my testing, the SmartThings integration worked better on Samsung phones compared to standard Android handsets, but your experience may vary.  If you are so inclined, you can create shopping lists directly on the fridge and then order them from partners including Instacart, Amazon Fresh, Walmart and others.  <h2 id="jump-link-wrap-up">Wrap-up</h2> AI in a fridge might sound ridiculous, but it allows for some genuinely useful features, even if the tech suffers from growing pains. And while this might sound preposterous, I think Samsung should actually double down on the tech by adding even more cameras, so the fridge has more angles and viewpoints to choose from as its algorithms evolve. The ability to automatically track what goes in and out of the fridge makes it easier to both shop and plan meals, and this is just in the early stages of the tech.  That said, obsessing over AI is sort of besides the point because even without it, Samsung's Bespoke French Door fridge is a first-class icebox. For tech-savvy people, its big screen helps simplify and organize our digital lives while simultaneously allowing the fridge to be helpful in ways you might not have even considered. It's also well-built and exquisitely designed to fit in seamlessly with other modern devices, which feels refreshing considering how often appliances feel like they are lagging years or decades behind gadgets like phones and laptops. And I can't overstate how much I like Samsung's Beverage Center, which in my opinion, is so good it's worth going with a Bespoke fridge just for that one feature.  In the end, the two biggest factors you'll need to consider are price and your appetite to digitize your kitchen, and by extension, your life. Starting at $3,500 for AI-equipped Bespoke models, Samsung's top-of-the-line fridges are far from cheap. Thankfully, versions without machine learning can be had for much less. But more importantly, you have to be willing to embrace technology's role in your home. You may not agree or want to pay this much for features like this just yet, but Samsung's Bespoke AI 4-Door French Door refrigerator has convinced me there's a happy home for smarter iceboxes. This article originally appeared on Engadget at https://www.engadget.com/home/kitchen-tech/samsung-bespoke-fridge-with-ai-review-all-the-bells-and-whistles-140000099.html?src=rss

Jan 23, 2026

Ring can now verify if a video has been altered

<a target="_blank" class="link" href="https://www.engadget.com/home/smart-home/ring-relaunches-its-suite-of-smart-home-sensors-140000667.html" data-i13n="cpos:1;pos:1">Ring</a> has launched a new tool that can tell you if a video clip captured by its camera has been altered or not. The company says that every video downloaded from Ring starting in December 2025 going forward will come with a digital security seal. “Think of it like the tamper-evident seal on a medicine bottle,” it explained. Its new tool, called <a target="_blank" class="link" href="https://blog.ring.com/about-ring/your-ring-videos-verified/" data-i13n="cpos:2;pos:1">Ring Verify</a>, can tell you if a video has been altered in any way. Even adjusting a video clip’s brightness or trimming a few seconds off will break that seal, and the tool will tell you that it cannot be verified. All you have to do is visit the <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=b50ccd99-4c85-48e0-a2ed-9c2b9b65422b&siteId=us-engadget&pageId=1p-autolink&contentUuid=9d4563e0-3185-47a5-a9a4-8c4221e5d167&featureId=text-link&merchantName=Ring&linkText=tool%E2%80%99s+web+page&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3JpbmcuY29tL3BhZ2VzL3ZlcmlmeSIsImNvbnRlbnRVdWlkIjoiOWQ0NTYzZTAtMzE4NS00N2E1LWE5YTQtOGM0MjIxZTVkMTY3Iiwib3JpZ2luYWxVcmwiOiJodHRwczovL3JpbmcuY29tL3BhZ2VzL3ZlcmlmeSJ9&signature=AQAAAUS4iAeOWt2tzx66SYE10UaF50uRahLMGdh46N64T8MV&gcReferrer=https%3A%2F%2Fring.com%2Fpages%2Fverify" data-i13n="elm:affiliate_link;sellerN:Ring;elmt:;cpos:3;pos:1" data-original-link="https://ring.com/pages/verify">tool’s web page</a> and upload the video you want to check. In the era of AI, you can also use the tool to make sure you’re looking at a real Ring video instead of something generated by artificial intelligence models. OpenAI’s <a target="_blank" class="link" href="https://www.engadget.com/ai/openai-now-sells-extra-sora-credits-for-4-plans-to-reduce-free-gens-in-the-future-223905628.html" data-i13n="cpos:4;pos:1">Sora</a>, for instance, can easily create Ring video doorbell footage with just a prompt or two, leading to a lot of fake security cam footage on social media apps. The company suggests asking the owner of the camera to share videos straight from the Ring app to make sure they’re unaltered. It also doesn’t matter what camera it is: All of the company’s models now embed digital security seals to downloaded videos and videos shared from Ring’s cloud storage. However, the tool can only tell you if a video has been tampered with or can’t be verified at all. It can’t tell you what has been edited or if it’s AI-generated. This article originally appeared on Engadget at https://www.engadget.com/home/smart-home/ring-can-now-verify-if-a-video-has-been-altered-140000140.html?src=rss

Jan 23, 2026

TurboTax Deluxe is on sale for only $45 ahead of tax season

With the spring approaching, it's time to start thinking about how you'll file your taxes. Unfortunately, with the <a data-i13n="cpos:1;pos:1" href="https://www.engadget.com/big-tech/the-irs-free-tax-filing-tool-reportedly-wont-be-offered-in-2026-185430716.html">recent demise of Direct File</a>, there isn't a free, government-provided alternative to TurboTax in 2026, and you'll need to look to other federal programs, including <a data-i13n="cpos:2;pos:1" href="https://www.irs.gov/e-file-do-your-taxes-for-free">IRS Free File</a>, to file your paperwork. If you don't qualify for Free File, chances are you'll need to dish out cash for Intuit's software. And while we hate to suggest you give money to a company that has <a data-i13n="cpos:3;pos:1" href="https://prospect.org/2025/04/17/2025-04-17-intuit-turbotax-wins-battle-against-taxpayers-irs-direct-file/">spent decades lobbying</a> to ensure America's tax system remains a nightmare, we can at least save you some money if TurboTax is your only option. <core-commerce id="91f223ca3a9046bc99a618db6550f3d3" data-type="product-list" data-original-url="https://www.amazon.com/TurboTax-Deluxe-Desktop-Federal-Download/dp/B0FWV56H48?"></core-commerce> Right now, Amazon has discounted the <a data-i13n="elm:affiliate_link;sellerN:Amazon;elmt:;cpos:4;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=66ea567a-c987-4c2e-a2ff-02904efde6ea&itemId=amazon_B0FWTS91PP&siteId=us-engadget&pageId=1p-autolink&contentUuid=8d4b28c6-34f1-44bb-800a-eec732a96ded&featureId=text-link&merchantName=Amazon&linkText=deluxe+desktop+edition+of+TurboTax&custData=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&signature=AQAAAW8L9G1hmZszyege3Mjq8IzopOwPoqZ8lfAwan_ZS8gi&gcReferrer=https%3A%2F%2Fwww.amazon.com%2FTurboTax-Premier-Desktop-Federal-Download%2Fdp%2FB0FWTS91PP%2Fref%3Dast_sto_dp_puis" class="rapid-with-clickid" data-original-link="https://www.amazon.com/TurboTax-Premier-Desktop-Federal-Download/dp/B0FWTS91PP/ref=ast_sto_dp_puis?th=1">deluxe desktop edition of TurboTax</a> by 44 percent. That gets you $35 off software that normally costs $80, which you can use to file both your individual state and federal taxes. For most Americans, that should cover their needs. However, if you're self-employed, you'll need to pony up for the more expensive <a data-i13n="elm:affiliate_link;sellerN:Amazon;elmt:;cpos:5;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=66ea567a-c987-4c2e-a2ff-02904efde6ea&itemId=amazon_B0FWV2NJV5&siteId=us-engadget&pageId=1p-autolink&contentUuid=8d4b28c6-34f1-44bb-800a-eec732a96ded&featureId=text-link&merchantName=Amazon&linkText=Business+version+of+TurboTax&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5hbWF6b24uY29tL1R1cmJvVGF4LUJ1c2luZXNzLURlc2t0b3AtRmVkZXJhbC1Eb3dubG9hZC9kcC9CMEZXVjJOSlY1L3JlZj1hc3Rfc3RvX2RwX3B1aXM_dGFnPWdkZ3QwYy0yMCIsImNvbnRlbnRVdWlkIjoiOGQ0YjI4YzYtMzRmMS00NGJiLTgwMGEtZWVjNzMyYTk2ZGVkIiwib3JpZ2luYWxVcmwiOiJodHRwczovL3d3dy5hbWF6b24uY29tL1R1cmJvVGF4LUJ1c2luZXNzLURlc2t0b3AtRmVkZXJhbC1Eb3dubG9hZC9kcC9CMEZXVjJOSlY1L3JlZj1hc3Rfc3RvX2RwX3B1aXMiLCJkeW5hbWljQ2VudHJhbFRyYWNraW5nSWQiOnRydWUsInNpdGVJZCI6InVzLWVuZ2FkZ2V0IiwicGFnZUlkIjoiMXAtYXV0b2xpbmsiLCJmZWF0dXJlSWQiOiJ0ZXh0LWxpbmsifQ&signature=AQAAAb0SqAIVO-2QeU6F8v1lz3jidDoYmdfaviPLW5VWaXo9&gcReferrer=https%3A%2F%2Fwww.amazon.com%2FTurboTax-Business-Desktop-Federal-Download%2Fdp%2FB0FWV2NJV5%2Fref%3Dast_sto_dp_puis" class="rapid-with-clickid" data-original-link="https://www.amazon.com/TurboTax-Business-Desktop-Federal-Download/dp/B0FWV2NJV5/ref=ast_sto_dp_puis?th=1">Business version of TurboTax</a>, which is currently 42 percent off, or $54 off. One last thing to note: you'll need a PC with either Windows 11 or macOS Sonoma to run the software, so an older machine won't do, unfortunately. Follow <a data-i13n="cpos:6;pos:1" href="https://twitter.com/EngadgetDeals">@EngadgetDeals</a> on X for the latest <a data-i13n="cpos:7;pos:1" href="https://www.engadget.com/deals/">tech deals</a> and <a data-i13n="cpos:8;pos:1" href="https://www.engadget.com/best-tech/">buying advice</a>.This article originally appeared on Engadget at https://www.engadget.com/deals/turbotax-deluxe-is-on-sale-for-only-45-ahead-of-tax-season-204848111.html?src=rss

Jan 23, 2026

Apple's AirPods Pro 3 drop back down to a record-low price

AirPods Pro 3 have returned to an all-time-low price. You can pick them up for $200 right now, which is $50 off and the best discount we've seen since the earbuds came out at the end of last year. We gave these AirPods a score of 90 out of 100 in our <a data-i13n="cpos:1;pos:1" href="https://www.engadget.com/audio/headphones/airpods-pro-3-review-a-significant-update-to-apples-best-earbuds-120028170.html">review</a> and we consider them the <a data-i13n="cpos:2;pos:1" href="https://www.engadget.com/audio/headphones/best-wireless-earbuds-120058222.html">best wireless earbuds for iPhone</a> users. The third generation model is a substantial update to an already impressive product, bringing new features like live translation to the table. <core-commerce id="9d2a9c264099463ba10d3ed1a8a4d49b" data-type="product-list" data-original-url="https://www.amazon.com/Apple-Cancellation-Translation-Headphones-High-Fidelity/dp/B0FQFB8FMG/?"></core-commerce> The AirPods Pro 3 also offer heart-rate tracking, allowing them to stream live data to Apple Fitness and other third-party workout apps. This lets you track your fitness and close activity rings similarly to when using an Apple Watch. Apple also improved the sound quality of this generation thanks in large part to a redesigned acoustic architecture that improves how air is delivered into the ear canal. Additionally, the company changed the angle of the drivers to help more sound get where it needs to go instead of bouncing off the sides of your ear. We felt the jump in audio quality from the last generation to this one was substantial. The ANC on the AirPods Pro 3 has also been improved. Apple claims this model filters out twice as much noise as the AirPods Pro 2 and four times as much as the original AirPods Pro. We found the noise cancellation to be top-notch. Live Translation is as neat as it sounds, leveraging the AirPods' noise cancellation tech, Apple's AI and beam-forming mics to help you understand a speaker in a different language. Right now, this is limited to a handful of languages with more being added over time. It's not often you see one of Apple's best and most recent products on sale for 20 percent off, so if you're in the market for an upgrade or your first set of AirPods, consider taking advantage of this deal. Follow <a data-i13n="cpos:3;pos:1" href="https://twitter.com/EngadgetDeals">@EngadgetDeals</a> on X for the latest <a data-i13n="cpos:4;pos:1" href="https://www.engadget.com/deals/">tech deals</a> and <a data-i13n="cpos:5;pos:1" href="https://www.engadget.com/best-tech/">buying advice</a>.This article originally appeared on Engadget at https://www.engadget.com/deals/apples-airpods-pro-3-drop-back-down-to-a-record-low-price-140752588.html?src=rss

Jan 23, 2026

Engadget Podcast: Why would Apple want to make an AI pin?

<a target="_blank" class="link" href="https://www.engadget.com/wearables/apple-is-reportedly-developing-a-wearable-ai-pin-204705065.html" data-i13n="cpos:1;pos:1">Apple is reportedly working on an AI pin</a> of its very own to compete with OpenAI's non-existent pin. No, it <a target="_blank" class="link" href="https://www.engadget.com/ai/an-ai-pin-is-beneath-apple-182744647.html" data-i13n="cpos:2;pos:1">doesn't really make much sense to us, either</a>. In this episode, Devindra and Engadget Managing Editor Cherlynn Low discuss why Apple might be quick to jump on the AI pin trend, even before it jumps into smart rings. Also, we dive into Sony’s Bravia TV deal with TCL, and the aftermath of Verizon’s massive service outage.<div><div><div style="left:0;width:100%;height:200px;position:relative;"><iframe src="https://iframely.publishing.yahoo.net/zWXYDgZL" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen data-embed-domain="player.simplecast.com" data-provider-name="iframe"></iframe></div></div></div><h2 id="jump-link-subscribe">Subscribe!</h2><ul><li><a target="_blank" class="linkDecorator_linkStyle__yM-Nh link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=9b2a3988-c5c9-4640-8e2b-1de7b2717343&siteId=us-engadget&pageId=1p-autolink&contentUuid=f69b37ef-2fc2-476c-9c75-96cba9a59b0e&featureId=text-link&merchantName=iTunes&linkText=iTunes&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL2l0dW5lcy5hcHBsZS5jb20vdXMvcG9kY2FzdC90aGUtZW5nYWRnZXQtcG9kY2FzdC9pZDExNDI3OTA1MzA_bXQ9MiIsImNvbnRlbnRVdWlkIjoiZjY5YjM3ZWYtMmZjMi00NzZjLTljNzUtOTZjYmE5YTU5YjBlIiwib3JpZ2luYWxVcmwiOiJodHRwczovL2l0dW5lcy5hcHBsZS5jb20vdXMvcG9kY2FzdC90aGUtZW5nYWRnZXQtcG9kY2FzdC9pZDExNDI3OTA1MzA_bXQ9MiJ9&signature=AQAAAS5N7PKPmfWjxsaLrNSnRidcs5wzr0hMUXiNZS0QFb7W&gcReferrer=https%3A%2F%2Fitunes.apple.com%2Fus%2Fpodcast%2Fthe-engadget-podcast%2Fid1142790530%3Fmt%3D2" data-i13n="elm:affiliate_link;sellerN:iTunes;elmt:;slk:iTunes;cpos:3;pos:1" data-original-link="https://itunes.apple.com/us/podcast/the-engadget-podcast/id1142790530?mt=2">iTunes</a></li><li><a target="_blank" class="linkDecorator_linkStyle__yM-Nh link rapid-with-clickid" href="https://open.spotify.com/show/3EC3dMIRB99dkvoSBeSO7G?si=f7UALGKmRDWtOrU8Z4-uiQ" data-i13n="slk:Spotify;elm:affiliate_link;sellerN:;elmt:;cpos:4;pos:1">Spotify</a></li><li><a target="_blank" class="linkDecorator_linkStyle__yM-Nh link" href="http://pca.st/mlY3" data-i13n="slk:Pocket Casts;cpos:5;pos:1">Pocket Casts</a></li><li><a target="_blank" class="linkDecorator_linkStyle__yM-Nh link" href="http://www.stitcher.com/podcast/engadget/the-engadget-podcast" data-i13n="slk:Stitcher;cpos:6;pos:1">Stitcher</a></li><li><a target="_blank" class="linkDecorator_linkStyle__yM-Nh link" href="https://podcasts.google.com/?feed=aHR0cDovL2ZlZWRzLnNvdW5kY2xvdWQuY29tL3VzZXJzL3NvdW5kY2xvdWQ6dXNlcnM6MjQ1MjIwOTU4L3NvdW5kcy5yc3M&ep=14" data-i13n="slk:Google Podcasts;cpos:7;pos:1">Google Podcasts</a></li></ul><h2 id="jump-link-topics">Topics</h2><ul><li>TCL is taking over Sony’s Bravia TV business in a new joint venture – 0:58</li><li>Last week’s Verizon outage was resolved after 10 hours, no official word on what caused it – 8:39</li><li>Youtube CEO promises more AI video tools for creators while also denouncing deepfakes – 12:19</li><li>The FTC isn’t giving up on its Meta antitrust case – 14:22</li><li>Trump family earnings from crypto may total $1.4B in 2025, but likely much more – 19:00</li><li>Adobe Acrobat can now generate presentations and podcasts from your documents – 21:12</li><li>Why the heck would Apple make an AI pin?? – 25:15</li><li>Around Engadget: Sony LinkBuds Clip review, Volvo EX60 and Canon EOS R6 III reviews – 43:14</li><li>Pop culture picks – 46:34 </li></ul><h2 id="jump-link-credits">Credits</h2>Host: Devindra Hardawar Guest: Cherlynn Low Producer: Ben Ellman Music: Dale North and Terrence O’BrienThis article originally appeared on Engadget at https://www.engadget.com/ai/engadget-podcast-why-would-apple-want-to-make-an-ai-pin-130732754.html?src=rss

Jan 23, 2026

The Morning After: Apple might be making its own AirTag-sized AI wearable

It’s been a while since rumors and reports suggested Apple is exploring a new divisive product category, and it’s been several years since the Apple car. Unfortunately, the new challenger is a wearable AI pin with cameras, mics and… <a target="_blank" class="link" href="https://www.engadget.com/wearables/apple-is-reportedly-developing-a-wearable-ai-pin-204705065.html" data-i13n="slk:zero interest from me;cpos:1;pos:1">zero interest from me</a>.According to a report from The Information, it’ll resemble a slightly thicker AirTag with an aluminum and glass exterior. The report suggests it’ll have two cameras (standard and wide-angle) for photos and video. It may also have three microphones and a (swoon) physical button. I love a physical button.<a href="https://www.engadget.com/wearables/apple-is-reportedly-developing-a-wearable-ai-pin-204705065.html"><figure><img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/a0572c40-c174-11f0-8fff-f232cbd8677f.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/a0572c40-c174-11f0-8fff-f232cbd8677f.jpg" style="height:396px;width:640px;" alt="TMA" data-uuid="10707724-3695-4e40-ba5e-3c25045e22cc"/><figcaption>TMA</figcaption><div class="photo-credit">Apple</div></figure></a>How is Apple going to pitch it? What is the non-creepy, not-nefarious selling point of a tiny listening device with cameras? As Engadget’s Devindra Hardawar puts it: Why attempt an unproven wearable? Especially when its own Watch and AirPods could already deliver a lot of what’s being rumored here. The Information says Apple could release its AI pin as early as 2027, but it’s reportedly only in the very early stages and could still be canceled. Even if the Vision Pro and Watch eventually happened, the Apple Car and the Apple TV never did.— Mat Smith<h2 id="jump-link-tiktok-finalizes-deal-for-its-us-entity"><a target="_blank" class="link" href="https://www.engadget.com/big-tech/tiktok-finalizes-deal-for-its-us-entity-010543484.html" data-i13n="slk:TikTok finalizes deal for its US entity;cpos:2;pos:1">TikTok finalizes deal for its US entity</a></h2>A group of non-Chinese investors will own 80 percent of it.TikTok owner ByteDance has finalized a deal for its US entity. The majority of its stake is held by a group of non-Chinese investors. The deal was closed just before the Trump administration’s latest deadline to ban the app in the US unless it was divested from ByteDance. TikTok’s new investors will own 80 percent, with Oracle, Silver Lake and MGX, an Emirati-state owned investment firm, taking 15 percent each.  According to TikTok’s announcement, the joint venture will protect American users’ data with Oracle’s secure US cloud environment. It will also retrain TikTok’s algorithm using US users’ data and will be responsible for content moderation in the US. The new US TikTok also promises interoperability, ensuring users still get international content and, if they’re creators, viewers.<a target="_blank" class="link" href="https://www.engadget.com/big-tech/tiktok-finalizes-deal-for-its-us-entity-010543484.html" data-i13n="slk:Continue reading.;cpos:3;pos:1">Continue reading.</a><h2 id="jump-link-bungies-marathon-arrives-on-march-5"><a target="_blank" class="link" href="https://www.engadget.com/gaming/bungies-marathon-arrives-on-march-5-193808588.html" data-i13n="slk:Bungie’s Marathon arrives on March 5;cpos:4;pos:1">Bungie’s Marathon arrives on March 5</a></h2>Another extraction shooter to distract.<figure><img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/b0d64230-fa76-11ed-a3ee-6a5e1adcfaa9.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/b0d64230-fa76-11ed-a3ee-6a5e1adcfaa9.jpg" style="height:427px;width:640px;" alt="TMA" data-uuid="ae4de90a-b84d-462b-8c23-70324e094547"/><figcaption>TMA</figcaption><div class="photo-credit">Bungie</div></figure>Bungie’s long-awaited Marathon will arrive on March 5. The 3v3 extraction shooter has a lot riding on it — but it looks pretty damn cool. You might remember Sony, Bungie’s parent company, previously committed to a September 2025 release. However, it delayed the game indefinitely last June after a mixed reception to its alpha and partially plagiarized visual assets.It’s a very important game for both Bungie and Sony. The latter said Destiny 2 had not lived up to its expectations. It wants another hit like Helldivers 2, not another <a target="_blank" class="link" href="https://www.engadget.com/gaming/bungies-marathon-arrives-on-march-5-193808588.html" data-i13n="slk:Concord;cpos:5;pos:1">Concord</a>.<a target="_blank" class="link" href="https://www.engadget.com/gaming/bungies-marathon-arrives-on-march-5-193808588.html" data-i13n="slk:Continue reading.;cpos:6;pos:1">Continue reading.</a><h2 id="jump-link-how-to-find-an-affordable-gpu-during-the-great-ramageddon"><a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/forza-horizon-6-will-hit-xbox-series-xs-and-pc-on-may-19-183242726.html" data-i13n="slk:How to find an affordable GPU during the great RAMageddon;cpos:7;pos:1">How to find an affordable GPU during the great RAMageddon</a></h2>It sucks out there.If you’re thinking about getting a new graphics card this year, your window for doing so at a typical retail price has closed. What do you do if you want to upgrade to a new graphics card this year? If you’re sitting on an older GPU, the best advice we can give is to stick with your current hardware. On the other hand, if your current GPU is not up to running the games you want to play, consider buying a card with at least 12GB of VRAM. Then, well, read on for more tips and our top recommendations.<a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/forza-horizon-6-will-hit-xbox-series-xs-and-pc-on-may-19-183242726.html" data-i13n="slk:Continue reading.;cpos:8;pos:1">Continue reading.</a><h2 id="jump-link-the-biggest-stories-you-might-have-missed">The biggest stories you might have missed</h2><ul><li><a target="_blank" class="link" href="https://www.engadget.com/apps/one-year-of-access-to-monarch-moneys-budgeting-app-is-down-to-50-right-now-204507836.html" data-i13n="slk:One year of access to Monarch Money’s budgeting app is down to $50 right now;cpos:9;pos:1">One year of access to Monarch Money’s budgeting app is down to $50 right now</a> </li><li><a target="_blank" class="link" href="https://www.engadget.com/gaming/fable-will-let-you-be-a-heartless-landlord-this-fall-200409310.html" data-i13n="slk:Fable will let you be a heartless landlord this fall;cpos:10;pos:1">Fable will let you be a heartless landlord this fall</a> </li><li><a target="_blank" class="link" href="https://www.engadget.com/ai/grok-generated-an-estimated-3-million-sexualized-images--including-23000-of-children--over-11-days-175053250.html" data-i13n="slk:Grok generated an estimated 3 million sexualized images — including 23,000 of children — over 11 days;cpos:11;pos:1">Grok generated an estimated 3 million sexualized images — including 23,000 of children — over 11 days</a> </li></ul>This article originally appeared on Engadget at https://www.engadget.com/general/the-morning-after-apple-might-be-making-its-own-airtag-sized-ai-wearable-121500060.html?src=rss

Jan 23, 2026

Apple accuses Europe of 'delay tactics' following alternative app store collapse

Following the shutdown of an alternative app store, Apple has accused the European Commission (EC) of using “political delay tactics” as an excuse to probe and fine the company, <a target="_blank" class="link" href="https://www.bloomberg.com/news/articles/2026-01-22/apple-hardware-chief-john-ternus-now-overseeing-design-tim-cook-ceo-succession" data-i13n="slk:Bloomberg;cpos:1;pos:1">Bloomberg</a> reported. Apple issued the statement preemptively as commission is reportedly preparing to blame Apple for the shutdown of third-party app store Setapp due to what the developer called “still-evolving and complex business terms.”As part of an EU ruling, Apple was forced to allow third-party marketplaces for apps once the Digital Markets Act (DMA) took effect in 2024. Apple agreed to allow such stores, but implemented fees of €0.50 per installation if the number of downloads exceeded one million, among other rules.In April 2025, the EC found Apple to be in “<a target="_blank" class="link" href="https://ec.europa.eu/commission/presscorner/detail/en/ip_25_1085" data-i13n="slk:non-compliance;cpos:2;pos:1">non-compliance</a>” with the DMA over rules on “steering” users to alternative payments and levied a <a target="_blank" class="link" href="https://www.engadget.com/big-tech/apple-and-meta-hit-with-first-fines-under-europes-new-digital-markets-act-112545713.html" data-i13n="slk:$500 million fine;cpos:3;pos:1">$500 million fine</a>. In June last year, Apple said it would <a target="_blank" class="link" href="https://www.engadget.com/apps/apple-details-new-fee-structures-for-app-store-payments-in-the-eu-204253948.html" data-i13n="slk:change its pricing;cpos:4;pos:1">change its pricing</a> for third-party App Stores to a five percent revenue share called the Core Technology Commission (CTC).However, Apple said that the EC has so far refused to allow those changes. In the meantime, a developer called MacPaw shut down its fledgling Setapp store, saying Apple’s terms “don’t fit Setapp’s current business model” due to the complexity. As a result, the EC will reportedly rule that Apple hasn’t addressed the key issues it raised regarding business terms, including their complexity.“The European Commission has refused to let us implement the very changes that they requested,” Apple said in a statement to Bloomberg. “In October, we submitted a formal compliance plan and they have yet to respond. The EC is using political delay tactics to mislead the public, move the goal posts, and unfairly target an American company with burdensome investigations and onerous fines.”In reply, an EC spokesperson told Engadget that it’s in “constant” contact with Apple with the aim of finding a DMA-compliant solution. “The Commission’s main objective is that gatekeepers operate in full compliance with the DMA. The Commission is available to discuss and work with gatekeepers to achieve full compliance. In this context, the Commission has been in a constant dialogue with Apple to solve all pending issues, while also listening to developers from all over the world. We remain committed to this exchange to find a DMA compliant solution.”This article originally appeared on Engadget at https://www.engadget.com/big-tech/apple-accuses-europe-of-delay-tactics-following-alternative-app-store-collapse-124701591.html?src=rss

Jan 23, 2026

The best GPS running watches for 2026

Having the right GPS watch on your wrist whether you’re going for your first ever run or your umpteenth run can make all the difference. The best GPS running watches not only keep track of how far you’ve run, but they track pace and other real-time metrics, advanced training features to help you hit your goals and, of course, precise distance measurements. Some models even provide offline maps for navigation, sleep tracking, recovery insights, and smart features that “regular” smartwatches do. For those who need extra durability and lasting battery life, higher-end sport watches — like some of the best Garmin watches — are built to handle intense workouts, harsh weather and long runs. If you're training for a marathon, triathlon or just want a multisport option that can keep up with your lifestyle, these watches have the tech to support you. With so many options available, from entry-level models to the best running watches packed with advanced running metrics, it can be tricky to find the right fit. That’s why we’ve rounded up our top picks to help you choose the perfect GPS watch for your training needs. <h2 id="jump-link-best-gps-running-watches-for-2026">Best GPS running watches for 2026</h2> <core-commerce id="c5dcc2fd49fb4f32bb536a40ed9d2cf0" data-type="product-list" data-original-url="https://www.garmin.com/en-US/p/1055469"></core-commerce> <core-commerce id="0ef73f5f54564aba9755f26458bdc3b3" data-type="product-list" data-original-url="https://www.amazon.com/dp/B0FQFPB851/ref=fs_a_wt2_us2?th=1"></core-commerce> <core-commerce id="63aae20affa345d08ed64cad4111178f" data-type="product-list" data-original-url="https://www.garmin.com/en-US/p/886725"></core-commerce> <core-commerce id="1845f99cd24f444a9b4c44991a0f52df" data-type="product-list" data-original-url="https://www.amazon.com/dp/B0CC62ZG1M?"></core-commerce> <h2 id="jump-link-other-gps-running-watches-we-tested">Other GPS running watches we tested</h2> <h3 id="jump-link-polar-pacer-pro">Polar Pacer Pro</h3> The <a data-i13n="elm:affiliate_link;sellerN:Amazon;elmt:;cpos:1;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=66ea567a-c987-4c2e-a2ff-02904efde6ea&itemId=amazon_B0B57M9ZKW&siteId=us-engadget&pageId=1p-autolink&contentUuid=5daa500c-3147-3dc9-8f5e-d508fa140873&featureId=text-link&merchantName=Amazon&linkText=Polar+Pacer+Pro&custData=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&signature=AQAAAUYf8oGLQYHtllG9NKohzJBl_ipvrtm8aSKc0DZiPQUL&gcReferrer=https%3A%2F%2Fwww.amazon.com%2FPolar-Pacer-Pro-Advanced-Ultra-Light%2Fdp%2FB0B57M9ZKW" class="rapid-with-clickid" data-original-link="https://www.amazon.com/Polar-Pacer-Pro-Advanced-Ultra-Light/dp/B0B57M9ZKW">Polar Pacer Pro</a> looked and felt quite similar to our top pick, and it mapped my outdoor runs accurately. However, Polar’s companion app is leagues behind Garmin’s with a confusing interface and a design that feels very much stuck in the past. It’s also $100 more expensive than our top pick. <h3 id="jump-link-amazfit-cheetah-pro">Amazfit Cheetah Pro</h3> The <a data-i13n="elm:affiliate_link;sellerN:Amazon;elmt:;cpos:2;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=66ea567a-c987-4c2e-a2ff-02904efde6ea&itemId=amazon_B0C7ZXCKK2&siteId=us-engadget&pageId=1p-autolink&contentUuid=5daa500c-3147-3dc9-8f5e-d508fa140873&featureId=text-link&merchantName=Amazon&linkText=Amazfit+Cheetah+Pro&custData=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&signature=AQAAAQFE1NxPXmn_c3yLVRhUIyWFzSNJ01h3_xY3jt72mEKQ&gcReferrer=https%3A%2F%2Fwww.amazon.com%2FAmazfit-Monitoring-Tracking-Water-Resistant-Bluetooth%2Fdp%2FB0C7ZXCKK2%2F" class="rapid-with-clickid" data-original-link="https://www.amazon.com/Amazfit-Monitoring-Tracking-Water-Resistant-Bluetooth/dp/B0C7ZXCKK2/">Amazfit Cheetah Pro</a> tracked my outdoor runs accurately and Zepp’s companion app has a coaching feature much like Garmin’s adaptive training plans that can outline a routine for you to complete in preparation for a race or to achieve a specific goal. My biggest issue with it was that its touchscreen wasn’t very responsive — it took multiple hard taps on the display to wake it, and often the raise-to-wake feature didn’t work, leaving me staring at a dark screen. <h2 id="jump-link-what-to-consider-before-buying-a-gps-running-watch">What to consider before buying a GPS running watch</h2> <h3 id="jump-link-gps-speed-and-accuracy">GPS speed and accuracy</h3> The most important thing for a GPS running watch to have is fast, accurate GPS tracking. That might seem obvious, but it’s quite easy to get distracted by all of the other smart features most of these devices have. Since most of them can be worn all day long as standard sport watches, there’s a lot of (possibly unnecessary) fluff that looks good on paper but won’t mean much if the core purpose if the device is left unfulfilled. To that end, I paid particular attention to how long it took each device’s built-in GPS tracking to grab my location before a run, if it ever lost my spot and the accuracy of the generated maps. Also, the device should be smart enough to let you start tracking a run while the GPS looks for your location. <h3 id="jump-link-workout-profiles-and-trackable-metrics">Workout profiles and trackable metrics</h3> You may not be able to suss out GPS accuracy just by looking at a spec sheet (that’s where this guide can help), but you can check for features like supported workout profiles. That’s something you’ll want to look into, even if your one and only activity is running. Check to make sure the best running watches you’re considering support all the kinds of running activities you like to do (outdoor runs, treadmill runs, etc) and any other workouts you may want to track with it. Most fitness wearables today aren’t one-trick ponies; you’ll find a healthy number of trackable exercise modes on any sport watch worth its salt. That said, the number of workout profiles can be directly proportional to a device’s price: the higher-end the product, chances are the more specific, precise workouts it can monitor. In a similar vein, you’ll want to check the trackable metrics of any watch you’re considering before you buy. Since we’re talking about the best GPS running watches, most will be able to track the basics like distance, heart rate and pace, and those are bare minimums. Some watches can monitor additional stats like speed, cadence, stride length, advanced running dynamics, aerobic and anaerobic training effect, intensity minutes and more. If you’re already a serious runner who trains for multiple races each year, or if you're a trail runner who needs elevation and navigation features, you’ll want to dig into the spec sheet of the watch you’re considering to make sure it can track all of your most necessary metrics. <h3 id="jump-link-size-and-weight">Size and weight</h3> It’s worth checking out a watch’s case size and weight before going all-in on one. GPS running watches, and standard smartwatches as well, can have a few different sizes to choose from so you’ll want to make sure you’re getting the best fit for your wrist. I have a smaller wrist, so I tend to avoid extra-large cases (anything over 42mm or so), especially if I intend on wearing the device all day long as my main timepiece. Weight, on the other hand, is a little less controllable, but typically smaller case sizes will save you a few grams in overall weight. For those who need durability, particularly trail runners or those tackling extreme conditions, devices like Garmin watches offer rugged builds that can handle rough terrain, impact, and extreme weather. <h3 id="jump-link-battery-life">Battery life</h3> Unlike regular smartwatches, GPS running watches have two types of battery life you’ll need to consider: with GPS turned on and in “smartwatch” mode. The former is more important than the latter because most GPS running watches have stellar battery life when used just as a smart timepiece. You can expect to get multiple days on a single charge, with some surviving more than two weeks (with all day and night wear) before they need a recharge. Battery life with GPS turned on will be much shorter by comparison, but any GPS running watch worth its salt should give you at least 10-15 hours of life with the GPS being used continuously. The more you’re willing to spend, the higher that number typically gets, with some GPS running watches lasting for 40 hours while tracking your location.This article originally appeared on Engadget at https://www.engadget.com/wearables/best-gps-running-watch-141513957.html?src=rss

Jan 22, 2026

Sennheiser introduces new TV headphones bundle with Auracast

Jan 22, 2026

Darth Maul's standalone series premieres on Disney+ on April 6

<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/oonACDq89BY?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Darth Maul, the beloved, sometimes spider-legged former Sith Lord first introduced in Star Wars: Episode 1 — The Phantom Menace, is officially getting his own animated spin-off on April 6, 2026, based on <a target="_blank" class="link" href="https://www.youtube.com/watch?v=oonACDq89BY" data-i13n="cpos:1;pos:1">a new teaser trailer</a> published by Disney. Star Wars: Maul — Shadow Lord was <a target="_blank" class="link" href="https://www.hollywoodreporter.com/tv/tv-news/darth-maul-animated-series-coming-to-disney-plus-1236194926/" data-i13n="cpos:2;pos:1">originally announced</a> at Star Wars Celebration in 2025, and is set after Maul's arc in Star Wars: The Clone Wars but before his appearance in Star Wars: Rebels.The 10-episode series covers Maul's plot to rebuild his criminal syndicate "on a planet untouched by the Empire," <a target="_blank" class="link" href="https://www.starwars.com/news/star-wars-maul-shadow-lord-first-trailer-poster-art" data-i13n="cpos:3;pos:1">according to Lucasfilm</a>. "There, he crosses paths with a disillusioned young Jedi Padawan who may just be the apprentice he is seeking to aid him in his relentless pursuit for revenge." Shadow Lord will premiere with two episodes on April 6, and will stream two episodes a week on Disney+ through May 6.Like most of Disney's popular animated Star Wars shows, Maul — Shadow Lord is created by Dave Filoni, a George Lucas supplicant, co-creator of The Clone Wars and <a target="_blank" class="link" href="https://www.engadget.com/entertainment/kathleen-kennedy-steps-down-as-lucasfilm-president-marking-a-new-era-for-the-star-wars-franchise-003407687.html" data-i13n="cpos:4;pos:1">newly minted</a> President and Chief Creative Officer at Luscafilm. Besides picking up the story threads he helped establish in his previous animated work, Maul — Shadow Lord could be representative of the work that will define his tenure: stories that play in the pre-existing Star Wars sandbox.Since completing their sequel trilogy in 2019 with Star Wars: The Rise of Skywalker, Disney and Lucasfilm have struggled to define what the future of their franchise will actually be. New projects have <a target="_blank" class="link" href="https://variety.com/2020/film/news/patty-jenkins-star-wars-1234851113/" data-i13n="cpos:5;pos:1">languished in pre-production</a>, interesting TV shows <a target="_blank" class="link" href="https://www.engadget.com/entertainment/tv-movies/star-wars-the-acolyte-isnt-getting-a-second-season-120033350.html" data-i13n="cpos:6;pos:1">have gotten cancelled</a> and the only things that seem to come out with any consistency are spin-offs of The Mandalorian and The Clone Wars. Disney and Lucasfilm will release <a target="_blank" class="link" href="https://www.engadget.com/entertainment/tv-movies/disneys-mandalorian-and-grogu-trailer-shows-the-fall-of-a-lumbering-giant-141515618.html" data-i13n="cpos:7;pos:1">The Mandalorian and Grogu</a> on May 22, 2026, the first new Star Wars movie to hit theaters in seven years. The Shawn Levy-directed <a target="_blank" class="link" href="https://www.engadget.com/entertainment/tv-movies/ryan-goslings-star-wars-movie-is-titled-starfighter-and-will-be-released-in-2027-123049184.html" data-i13n="cpos:8;pos:1">Star Wars: Starfighter</a>, the first film set chronologically after the events of Rise of Skywalker, is slated to premiere on May 28, 2027.This article originally appeared on Engadget at https://www.engadget.com/entertainment/streaming/darth-mauls-standalone-series-premieres-on-disney-on-april-6-224036720.html?src=rss

Jan 22, 2026

JBL made a pair of AI-powered practice amps

JBL is trying its hand at something new, with a pair of AI-powered practice amps. The BandBox Solo and BandBox Trio include an onboard Stem AI that purportedly lets you separate or remove vocals and instruments from any music streamed over Bluetooth. So, say you're a young guitarist learning “Stairway to Heaven” (as one does). At least in theory, you could use the speaker to remove Jimmy Page's part and hone your chops with the rest of the band.The $250 BandBox Solo, designed for individual musicians, has a single guitar / mic input. And the $600 BandBox Trio, better for bands, supports up to four instrument inputs. The latter also has a few extra perks not found in the Solo: onboard controls (to reduce reliance on the app) and an LCD.<figure><img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/jbl.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/jbl.jpg" style="height:1079px;width:1915px;" alt="A hand adjusting knobs on the JBL BandBox Trio speaker" data-uuid="f3522b07-5c29-4542-8521-4dcd3a0b253d"><figcaption>The more expensive BandBox Trio has an LCD and more onboard controls.</figcaption><div class="photo-credit">JBL</div></figure>The JBL One app lets you add filters to match a variety of modern and vintage amp models. You'll also find effects like phaser, chorus, and reverb, as well as a pitch shifter and tuner. A built-in looper will allow for layering, but JBL says that feature will arrive via a software update in October.JBL says the BandBox Solo supports up to six hours of battery life. The BandBox Trio is said to last up to 10 hours. But the latter has a replaceable battery, so you could buy a spare (for an as-yet-unknown price) and double that time.The <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=f671e4c0-78f5-4be4-a1dc-bc74e91d5c55&siteId=us-engadget&pageId=1p-autolink&contentUuid=b2a5cbfe-8d8c-4131-b0b7-d3db14c557cc&featureId=text-link&merchantName=JBL&linkText=%24250+BandBox+Solo&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5qYmwuY29tL0JBTkRCT1gtU09MTy5odG1sIiwiY29udGVudFV1aWQiOiJiMmE1Y2JmZS04ZDhjLTQxMzEtYjBiNy1kM2RiMTRjNTU3Y2MiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3LmpibC5jb20vQkFOREJPWC1TT0xPLmh0bWwifQ&signature=AQAAAfm1YIEn_qn3itz64ngtL8RHagmhk9uoGtW3-Ny8j-nG&gcReferrer=https%3A%2F%2Fwww.jbl.com%2FBANDBOX-SOLO.html" data-i13n="elm:affiliate_link;sellerN:JBL;elmt:;cpos:1;pos:1" data-original-link="https://www.jbl.com/BANDBOX-SOLO.html">$250 BandBox Solo</a> and <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=f671e4c0-78f5-4be4-a1dc-bc74e91d5c55&siteId=us-engadget&pageId=1p-autolink&contentUuid=b2a5cbfe-8d8c-4131-b0b7-d3db14c557cc&featureId=text-link&merchantName=JBL&linkText=%24600+BandBox+Trio&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5qYmwuY29tL0JBTkRCT1gtVFJJTy5odG1sIiwiY29udGVudFV1aWQiOiJiMmE1Y2JmZS04ZDhjLTQxMzEtYjBiNy1kM2RiMTRjNTU3Y2MiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3LmpibC5jb20vQkFOREJPWC1UUklPLmh0bWwifQ&signature=AQAAAfgztluLlU4JOEbPLhBkKYp0y-0fnc6onVPT_eGbZVUI&gcReferrer=https%3A%2F%2Fwww.jbl.com%2FBANDBOX-TRIO.html" data-i13n="elm:affiliate_link;sellerN:JBL;elmt:;cpos:2;pos:1" data-original-link="https://www.jbl.com/BANDBOX-TRIO.html">$600 BandBox Trio</a> are available to pre-order from JBL's website starting today. Shipments and third-party retail availability are scheduled for March 1.This article originally appeared on Engadget at https://www.engadget.com/audio/speakers/jbl-made-a-pair-of-ai-powered-practice-amps-221000631.html?src=rss

Jan 22, 2026

Telly has only delivered 35,000 of its free televisions with always-on ads

Jan 22, 2026

David Ellison extends deadline for Warner Bros. Discovery takeover offer

Paramount Skydance CEO David Ellison is apparently still hopeful that investors will approve his $108.4 billion hostile takeover of Warner Bros. Discovery. Paramount Skydance <a target="_blank" class="link" href="https://www.paramount.com/press/paramount-files-proxy-materials-and-extends-tender-offer" data-i13n="cpos:1;pos:1">announced Thursday</a> that it's extending its all-cash offer to acquire the storied studio, and giving investors until February 20, 2026 to accept. The company's previous offer expired on January 21, but with <a target="_blank" class="link" href="https://www.engadget.com/entertainment/paramount-wont-quit-files-suit-against-warner-bros-discovery-over-rejected-bid-175317166.html" data-i13n="cpos:2;pos:1">a lawsuit in the works</a> and <a target="_blank" class="link" href="https://www.wbd.com/news/netflix-and-warner-bros-discovery-amend-agreement-all-cash-transaction" data-i13n="cpos:3;pos:1">a revised Netflix deal</a> to compete with, Paramount Skydance wants to stay in the conversation.Netflix and Warner Bros. Discovery originally announced their <a target="_blank" class="link" href="https://www.engadget.com/entertainment/netflix-to-buy-warner-bros-for-827-billion-120836295.html" data-i13n="cpos:4;pos:1">$82.7 billion acquisition agreement</a> in December 2025. Netflix's deal is for a significant portion, but notably not all, of Warner Bros. Discovery as it exists today. If approved, the streaming service would acquire Warner Bros. film studios, New Line Cinema, HBO, HBO Max, the company's theme parks, game studios and select linear channels like TNT, but not the collection of reality TV and news programming that Warner Bros. Discovery calls <a target="_blank" class="link" href="https://www.engadget.com/entertainment/warner-bros-discovery-will-split-into-two-media-giants-144553585.html" data-i13n="cpos:5;pos:1">“Global Networks.”</a>Paramount Skydance made its <a target="_blank" class="link" href="https://ir.paramount.com/news-releases/news-release-details/paramount-launches-all-cash-tender-offer-acquire-warner-bros" data-i13n="cpos:6;pos:1">competing offer</a> of $108.4 billion for all of Warner Bros. Discovery a few days later in December, with the recommendation that shareholders reject the Netflix deal. To add pressure, Paramount Skydance also <a target="_blank" class="link" href="https://www.engadget.com/entertainment/paramount-wont-quit-files-suit-against-warner-bros-discovery-over-rejected-bid-175317166.html" data-i13n="cpos:7;pos:1">sued</a> Warner Bros. Discovery in January alleging that the company had not provided adequate information about why it favored Netflix over Paramount. Beyond offering more money, Paramount contends its deal is more likely to be approved by regulators because owning Warner Bros. doesn't "entrench Netflix's market dominance." <a target="_blank" class="link" href="https://s201.q4cdn.com/336605034/files/doc_downloads/2026/FINAL-WBD-Amended-14D-9-Shareholder-Letter-01-07-25.pdf" data-i13n="cpos:8;pos:1">Warner Bros. Discovery claims that</a> funding for Paramount's deal "remains inadequate" and that the company is uncertain Paramount Skydance will actually be able to complete the deal.David Ellison was previously able to <a target="_blank" class="link" href="https://www.engadget.com/paramount-and-skydance-will-merge-to-create-new-tech-media-giant-140052942.html" data-i13n="cpos:9;pos:1">merge Skydance with Paramount </a>using the financial backing of his billionaire father Larry Ellison, and the Ellison family's friendly relationship with the Trump administration. Promising to make sure that CBS News represents "a diversity of viewpoints” via a newly appointed <a target="_blank" class="link" href="https://deadline.com/2025/08/skydance-paramount-merger-cbs-news-1236481168/" data-i13n="cpos:10;pos:1">ombudsman</a>, and that the merged Paramount Skydance won't create any diversity, equity and inclusion programs was enough <a target="_blank" class="link" href="https://www.engadget.com/entertainment/fcc-approves-skydances-8-billion-paramount-acquisition-032028104.html" data-i13n="cpos:11;pos:1">to get the FCC to approve the merger</a>. Ellison might have thought acquiring Warner Bros. Discovery would be equally easy, but at least so far that hasn't worked out as planned.This article originally appeared on Engadget at https://www.engadget.com/entertainment/david-ellison-extends-deadline-for-warner-bros-discovery-takeover-offer-204752313.html?src=rss

Jan 22, 2026

Fable will let you be a heartless landlord this fall

After half a decade of waiting, Microsoft finally offered an <a target="_blank" class="link" href="https://news.xbox.com/en-us/2026/01/22/fable-interview-overview-details-developer-direct-2026/" data-i13n="cpos:1;pos:1">extended preview</a> of its long-awaited reboot of the beloved Fable RPG franchise on Thursday. During the company’s <a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/xbox-developer-direct-2026-how-to-watch-and-what-we-expect-to-see-from-fable-forza-horizon-6-and-beast-of-reincarnation-161000242.html" data-i13n="slk:recent Developer_Direct showcase;cpos:2;pos:1">recent Developer Direct showcase</a>, Playground Games, best known for its works on the Forza Horizon series, shared nearly 11 minutes of gameplay footage and commentary related to the upcoming title. The new game will feature a fully open world for players to explore, with locations from previous Fable games like Bowerstone making a return. <div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/C5qWGaqzk98?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Playground didn’t say exactly where the new game takes place in the Fable timeline. If you played 2010’s Fable 3, you may recall there was a time gap between each of the original games, with the final Lionhead entry giving players the chance to explore a version of Albion that’s been thrown into the tumult of an industrial revolution. This new game appears to offer a more traditional Medieval fantasy setting. From what little Playground shared of the story, the player character is the first hero born in Albion in a generation. Like previous Fable games, their story will begin when they’re a child. They’ll leave their home of Briar Hill after a “mysterious stranger” turns all the inhabitants into stone. Before you set off on your adventure, you’ll be able to decide what your hero looks like, with customization including options for different skin tones, head shapes, tattoos and scars. All pretty standard stuff, and combat doesn’t look groundbreaking. Your character can use a combination of melee and ranged weapons, alongside magic, to defeat their foes. In addition to a health bar, enemies appear to have a posture meter players can break through both light and heavy attacks, before carrying out “stylish” finishers. Judging from the footage Playground shared, combat doesn’t quite have the kinetic feel of say a FromSoftware title. <div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/IvB6AGwEINs?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Thankfully, there’s more to the game than just fighting monsters. Like past Fable games, players can ask villagers out on dates, before eventually marrying them and having children, and you’re not limited to a monogamous relationship. It looks like Playground built a modern polycule simulator. Beyond that, you can also purchase every house and business in Albion. The game’s NPCs will develop an opinion of your hero and their actions. Say you evict an artist that lived in one of the houses you bought, you can later meet them on the street and they’ll tell you to your face that you’re awful. Wonderful stuff. From a technical perspective, the footage Playground showed off looked on the rough side. The game’s frame rate and frame pacing wasn’t smooth, and I’m pretty sure I saw ghosting in some of the animations. Hopefully the studio has enough time between now and when the game is finally released sometime this autumn to polish the presentation. When the game does arrive, it will be available on Xbox Series X/S, PC via Xbox and Steam, PlayStation 5 and Game Pass Ultimate. Microsoft first announced a new Fable game was in development <a target="_blank" class="link" href="https://www.engadget.com/fable-xbox-series-x-windows-10-playground-games-confirmed-171804707.html" data-i13n="cpos:3;pos:1">back in 2020.</a> It then went three years before <a target="_blank" class="link" href="https://www.engadget.com/take-a-peek-at-xboxs-fable-reboot-in-action-172438277.html" data-i13n="cpos:4;pos:1">sharing a preview of the title</a>, which was originally slated to arrive in 2025 before it was <a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/xbox-has-delayed-its-fable-reboot-until-2026-173118802.html" data-i13n="cpos:5;pos:1">delayed to this year last February</a>. This article originally appeared on Engadget at https://www.engadget.com/gaming/fable-will-let-you-be-a-heartless-landlord-this-fall-200409310.html?src=rss

Jan 22, 2026

Double Fine announces delightful-looking multiplayer pottery game Kiln

<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/mi0iRV0U3VY?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Double Fine has been on a tear with its smaller projects lately. The popular indie game developer is following up last year's atmospheric adventure game <a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/doublefine-keeper-gameplay-gamescom2025-123038715.html" data-i13n="cpos:1;pos:1">Keeper</a> with a new title in a totally different direction. As its Double Fine founder Tim Schaefer attempted to say five times fast during today's Xbox Developer Direct, <a target="_blank" class="link" href="https://www.doublefine.com/games/kiln" data-i13n="cpos:2;pos:1">Kiln</a> is an "online multiplayer pottery party brawler" that boasts exactly the sort of colorful, clever fun that fans have come to expect from the studio. Kiln is a game with two facets: creation and destruction. On the creation side, players get to design a ceramic vessel of their choosing, anything from a dainty saucer to a massive vat. From the clips showed, this riff on character creation imitates many of the real processes of throwing on the wheel. There are options to decorate your vessel with glazes and stickers for additional personalization.Once you've made your pottery creation, it's time to destroy it. The game mode that Double Fine showed off during the Developer Direct is called Quench. In these capture-the-flag style matches, your teammates are trying to carry water in the vessels they've designed to the enemy's kiln, where the goal is to douse its flames before opponents can do the same to yours. Different vessel types have different attacks for smashing enemy ceramics to shards, so there can be a strategic angle to building up a smart team composition with a mix of defensive and offensive capabilities.Kiln looks like a really fun time, and it is due out in spring 2026. The game will be available on Xbox Series X/S, PlayStation 5 and Steam, as well as through Xbox programs such as Game Pass Ultimate and Xbox Cloud Gaming. This article originally appeared on Engadget at https://www.engadget.com/gaming/double-fine-announces-delightful-looking-multiplayer-pottery-game-kiln-195837155.html?src=rss

Jan 22, 2026

Substack now has a TV app

It was only 16 months ago that Substack <a target="_blank" class="link" href="https://www.engadget.com/apps/substack-a-newsletter-service-is-now-also-a-live-video-company-150004585.html" data-i13n="cpos:1;pos:1">expanded into live video</a>. Since then, it's become a widely adopted medium on the platform, including by its ever-growing <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=c813ae39-7d58-41cb-ac66-ad830606ceef&siteId=us-engadget&pageId=1p-autolink&contentUuid=cad4782e-89f8-4401-b62a-617b27bcfec7&featureId=text-link&merchantName=The+New+York+Times&linkText=list&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5ueXRpbWVzLmNvbS8yMDI1LzAzLzMxL2J1c2luZXNzL21lZGlhL3R2LW5ld3MtYW5jaG9ycy1zdWJzdGFjay5odG1sIiwiY29udGVudFV1aWQiOiJjYWQ0NzgyZS04OWY4LTQ0MDEtYjYyYS02MTdiMjdiY2ZlYzciLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3Lm55dGltZXMuY29tLzIwMjUvMDMvMzEvYnVzaW5lc3MvbWVkaWEvdHYtbmV3cy1hbmNob3JzLXN1YnN0YWNrLmh0bWwifQ&signature=AQAAAds2T5x6fgycGvIQxqk9Jd9VugThpHQFurl0KQ4z3d-v&gcReferrer=https%3A%2F%2Fwww.nytimes.com%2F2025%2F03%2F31%2Fbusiness%2Fmedia%2Ftv-news-anchors-substack.html" data-i13n="elm:affiliate_link;sellerN:The New York Times;elmt:;cpos:2;pos:1" data-original-link="https://www.nytimes.com/2025/03/31/business/media/tv-news-anchors-substack.html">list</a> of cable news expats. Now, the company has <a target="_blank" class="link" href="https://on.substack.com/p/introducing-the-substack-tv-app-now" data-i13n="cpos:3;pos:1">decided</a> it's ready for a bigger screen. Starting today, you can install a new Substack TV app on Apple TV and Google TV devices.Substack describes the app's initial launch as focusing on "reliable, high-quality viewing" for the platform's long-form videos. It centers around video posts and livestreams from the creators and publications you follow. There's also a "For You" section that combines those videos with algorithmic recommendations. And you can browse dedicated pages for each channel.Viewing access aligns with your subscription level. So, if a channel restricts videos to paying subscribers on Substack’s mobile app and website, the same paywall applies here. However, the company says it's working on adding previews of paid content for free subscribers. It also plans to add audio posts, search, improved discovery and separate sections for different shows within a single publication.At least one of those legacy media expats who moved to Substack is optimistic about the feature. "This is a game-changing moment for the rise of independent media," former CNN anchor Jim Acosta wrote in the announcement post. "Substack has proven that legacy media consumers are not only searching for fresh alternatives; they are finding them."Based on the comments in Substack’s announcement, some of the platform’s die-hards are less enthusiastic than Acosta. “File this under – thing we didn’t ask for,” Ashli Pollard wrote. “This is not YouTube,” a commenter named Dustin added. “Elevate the written word.” I personally don’t see any harm in another way to watch the videos Substack already hosts. (Its newsletters aren’t going anywhere.) But given how Big Tech companies tend to pivot further toward the most en vogue (and profitable) medium of the day, I can see why purists feel the need to dig in their heels.Regardless, you can find the Substack TV app by searching for its name in the tvOS App Store or the Apps section on Google and Android TV devices. You can sign in by scanning a QR code with your phone.This article originally appeared on Engadget at https://www.engadget.com/entertainment/streaming/substack-now-has-a-tv-app-195408592.html?src=rss

Jan 22, 2026

Elon Musk just told Davos that Tesla will sell humanoid robots next year, really, he swears

Elon Musk just took the stage at the World Economic Forum in Davos, Switzerland and announced that Tesla's Optimus humanoid robot <a data-i13n="cpos:1;pos:1" href="https://www.youtube.com/watch?v=SQJIyR5uoZY"><ins>will be sold to the public</ins></a> by the end of next year. Musk is the <a data-i13n="cpos:2;pos:1" href="https://en.wikipedia.org/wiki/List_of_predictions_for_autonomous_Tesla_vehicles_by_Elon_Musk"><ins>master of unrealistic timetables</ins></a>, but this may be the nuttiest one yet. These are humanoid robots that are supposed to be able to do just about any task a human can do. Musk, as usual, gave himself an out if the robots don't start rolling off the assembly line in 2027, saying that they'll only be released when Tesla is "confident that it’s very high reliability, very high safety and the range of functionality is also very high." <div id="9625071a62cd4e7a8e577a326b14a004"><iframe width="560" height="315" src="https://www.youtube.com/embed/SQJIyR5uoZY?si=OOkWUbFttS1XZCz3" title="YouTube video player" frameborder="0" allowfullscreen></iframe></div> He stated that the robots have already begun doing simple tasks in the Tesla factory, but there's no proof of this other than his word. In the real world, Optimus robots have continuously failed to live up to the marketing hype.  <div id="144acde4d07f4aa3884beee84959b622"><blockquote class="twitter-tweet">If there was any question that Optimus uses teleop for their robots. Here one clearly has a guy take the headset off and it falls over. Absolutely hilarious though. <a href="https://t.co/4gYVohjY00">pic.twitter.com/4gYVohjY00</a>— CIX 🦾 (@cixliv) <a href="https://twitter.com/cixliv/status/1997878834956525898?ref_src=twsrc%5Etfw">December 8, 2025</a></blockquote> </div> There have been plenty of reports suggesting that previous demos of the robots in action were actually smoke and mirrors, as they were being <a data-i13n="cpos:3;pos:1" href="https://finance.yahoo.com/news/tesla-optimus-bots-were-controlled-214754156.html"><ins>remotely piloted by human operators</ins></a>. Tesla's stock jumped over three percent upon this announcement because company investors live in a world of magical rainbows, unicorns and robots mixing drinks at home in the year 2027. Musk also recently stated that the robots would be ready for commercial deployment in 2026. That's this year. It's worth noting that the program head for the project, Milan Kovac, <a data-i13n="cpos:4;pos:1" href="https://www.engadget.com/big-tech/the-head-of-teslas-optimus-humanoid-robot-program-leaves-the-company-120015282.html"><ins>recently left the company</ins></a>. Again, these are autonomous robots that are supposed to be able to perform complex tasks across a wide variety of categories. That'll likely happen someday, <a data-i13n="cpos:5;pos:1" href="https://www.axios.com/2026/01/07/ai-nvidia-tesla-optimus-humanoid-robots"><ins>but not by 2027</ins></a>. I'm just gonna go ahead and throw Musk's timeline in with "two years to AGI" and "five years until the singularity." <div id="b3bbdec3b4974514ada8dd60fc8560da"><iframe width="560" height="315" src="https://www.youtube.com/embed/Qfj4urMF8CU?si=YNotLdBnBvYoomY7" title="YouTube video player" frameborder="0" allowfullscreen></iframe></div> On a related note, the Tesla CEO also said that the long-awaited Cybercab will enter production in April, with a goal of <a data-i13n="cpos:6;pos:1" href="https://timesofindia.indiatimes.com/technology/tech-news/elon-musk-shares-an-update-on-teslas-cybercab-robotaxi-and-optimus-robot-production-initially-it-will-be/articleshow/127179902.cms"><ins>two million vehicles manufactured each year</ins></a>. This isn't quite as far-fetched as the Optimus promise, but that two million number seems highly suspect. Just how many consumers does he think want a car without a steering wheel that only holds two people?This article originally appeared on Engadget at https://www.engadget.com/big-tech/elon-musk-just-told-davos-that-tesla-will-sell-humanoid-robots-next-year-really-he-swears-192515568.html?src=rss

Jan 22, 2026

Mario Kart World adds a team option in Knockout Tour

Jan 22, 2026

Forza Horizon 6 will hit Xbox Series X/S and PC on May 19

<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/pWw-UENvdTw?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Forza Horizon 6 is taking the racing series to Japan for the first time, and now we know exactly when you’ll be able to hit the virtual streets of Tokyo. The game will <a target="_blank" class="link" href="https://news.xbox.com/en-us/2026/01/22/forza-horizon-6-developer-direct-breakdown-interview/" data-i13n="cpos:1;pos:1">arrive</a> on Xbox Series X/S, PC and Xbox Cloud on May 19, with an early access period commencing on May 15. That lines up with <a target="_blank" class="link" href="https://www.videogameschronicle.com/news/forza-horizon-6s-release-date-may-have-been-revealed-early-by-an-in-game-ad/" data-i13n="cpos:2;pos:1">earlier reports</a> about the release date(s) after an in-game ad supposedly appeared in Forza Horizon 5. The upcoming title will hit PS5 later this year.The standard edition of Forza Horizon 6 is <a target="_blank" class="link" href="https://www.dealabs.com/magazine/forza-horizon-6-oui-une-version-physique-sera-bien-prevue-a-cette-date-et-a-ce-prix-on-vous-revele-les-details-du-lancement-60887" data-i13n="cpos:3;pos:1">expected</a> to cost $70, though it will be on Xbox Game Pass Ultimate and PC Game Pass. To jump in a few days early, you’ll need the premium edition (a deluxe edition will be available too).Playground Games and co-developer Turn 10 Studios offered a first look at gameplay during Thursday’s <a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/xbox-developer-direct-2026-how-to-watch-and-what-we-expect-to-see-from-fable-forza-horizon-6-and-beast-of-reincarnation-161000242.html" data-i13n="cpos:4;pos:1">Xbox Developer Direct showcase</a>. You’ll start out by working through Horizon qualifiers in order to join the Horizon Festival racing circuit, then you’ll unlock faster cars and other goodies as you rise up through the ranks. In a few feature called Horizon Rush, you’ll take on obstacle courses. Playground says this is the largest Forza Horizon game to date. This version of Tokyo, for instance, is said to be five times larger than any urban area the studio has built before and it has multiple districts. There are lots of rural areas in the game as well, and the environments are affected by seasonal changes. As has been the case in previous games, you’ll be able to explore the world as you wish, either solo or with your friends.There will be more than 550 cars available at the jump, and you’ll be able to apply liveries to their windows. Each player house has a customizable garage. You can build up a part of the world called The Estate as you see fit and your friends can visit to see what you’ve made (building will be available in multiplayer for the first time in the series too). There are also new open-world car meets where you can show off your whips, buy copies of other players’ rides and download more songs.Playground is including all of the accessibility features from previous games. It’s adding new ones as well, such as a customizable high-contrast mode and an autodrive option.This article originally appeared on Engadget at https://www.engadget.com/gaming/xbox/forza-horizon-6-will-hit-xbox-series-xs-and-pc-on-may-19-183242726.html?src=rss

Jan 22, 2026

An AI pin is beneath Apple

So it's come to this: Apple is <a target="_blank" class="link" href="https://www.engadget.com/wearables/apple-is-reportedly-developing-a-wearable-ai-pin-204705065.html" data-i13n="cpos:1;pos:1">reportedly working on a wearable AI pin</a>. According to <a target="_blank" rel="" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=ba0a4cdc-cec8-416a-9e93-e11b8179129c&siteId=us-engadget&pageId=1p-autolink&contentUuid=6a9c4806-23e0-4b7e-8496-a6afa6f6c483&featureId=text-link&merchantName=The+Information&linkText=The+Information&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy50aGVpbmZvcm1hdGlvbi5jb20vYXJ0aWNsZXMvYXBwbGUtZGV2ZWxvcGluZy1haS13ZWFyYWJsZS1waW4iLCJjb250ZW50VXVpZCI6IjZhOWM0ODA2LTIzZTAtNGI3ZS04NDk2LWE2YWZhNmY2YzQ4MyIsIm9yaWdpbmFsVXJsIjoiaHR0cHM6Ly93d3cudGhlaW5mb3JtYXRpb24uY29tL2FydGljbGVzL2FwcGxlLWRldmVsb3BpbmctYWktd2VhcmFibGUtcGluIn0&signature=AQAAAaDFk1Yg79gW6Rg5Dq3rtuPYwVmJLtSGdSE3fFJZja-o&gcReferrer=https%3A%2F%2Fwww.theinformation.com%2Farticles%2Fapple-developing-ai-wearable-pin" data-i13n="elm:affiliate_link;sellerN:The Information;elmt:;cpos:2;pos:1" data-original-link="https://www.theinformation.com/articles/apple-developing-ai-wearable-pin">The Information</a>, it is going to be a small device with "multiple cameras, a speaker, microphones and wireless charging." It sounds like the perfect gadget to pair with the <a target="_blank" class="link" href="https://www.engadget.com/ai/apples-ai-infused-siri-may-not-show-up-until-spring-2026-221212681.html" data-i13n="cpos:3;pos:1">long-awaited AI-powered Siri update</a>, which will also reportedly <a target="_blank" class="link" href="https://www.engadget.com/ai/apple-is-reportedly-overhauling-siri-to-be-an-ai-chatbot-205303818.html" data-i13n="cpos:4;pos:1">work as a chatbot</a>. But while many Apple rumors conjure up an air of excitement, the notion of an Apple AI pin sounds downright baffling. Worse, it just seems desperate. Apple, the company known for taking its time to jump into new categories with more thoughtful solutions than its competitors, is reportedly chasing the specter of OpenAI's unreleased AI pin. Never mind that OpenAI has never actually produced any hardware, and that it arguably stumbled into its position as a leading AI player. And never mind that <a target="_blank" class="link" href="https://www.engadget.com/the-humane-ai-pin-is-the-solution-to-none-of-technologys-problems-120002469.html" data-i13n="cpos:5;pos:1">Humane's AI pin was a notorious failure</a> that barely worked, and seemed pointless from the start. <div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/1JjInHnLmGI?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Sure, Apple doesn't want more AI eggs on its face, after the delay of its Siri revamp and the underwhelming (and error-prone) <a target="_blank" class="link" href="https://www.engadget.com/ai/to-fix-apple-intelligence-apple-needs-to-be-honest-about-its-capabilities-130046256.html" data-i13n="cpos:6;pos:1">debut of Apple Intelligence</a>. Beyond OpenAI, there's also competition from Meta's Ray-Ban smart glasses, which lean heavily on the company’s AI. There’s also the looming threat of whatever AI hardware Meta is cooking up next, following the <a target="_blank" class="link" href="https://www.engadget.com/ar-vr/meta-refocuses-on-ai-hardware-as-metaverse-layoffs-begin-145924706.html" data-i13n="cpos:7;pos:1">layoffs from its virtual reality division</a>. And while Google doesn’t have much to show from its <a target="_blank" class="link" href="https://www.engadget.com/ar-vr/google-xr-glasses-hands-on-lightweight-but-with-a-limited-field-of-view-213940554.html" data-i13n="cpos:8;pos:1">Android XR platform</a>, which aims to bring its Gemini AI to your face, <a target="_blank" class="link" href="https://www.engadget.com/ar-vr/google-and-samsungs-first-android-xr-headset-is-the-1800-galaxy-xr-020004449.html" data-i13n="cpos:9;pos:1">Samsung’s Galaxy XR</a> is a start. We’ve also recently seen compelling demos of <a target="_blank" class="link" href="https://www.engadget.com/wearables/heres-how-google-is-laying-the-foundation-for-our-mixed-reality-future-180000716.html" data-i13n="cpos:10;pos:1">Google’s AR glasses prototypes and Xreal’s Project Aura glasses</a>.If Apple's AI pin serves as a conduit to Siri, is it really that much more convenient than using an iPhone, AirPods or even an Apple Watch to do the same? The company has <a target="_blank" class="link" href="https://www.bloomberg.com/news/articles/2025-05-22/apple-plans-glasses-for-2026-as-part-of-ai-push-nixes-watch-with-camera?embedded-checkout=true" data-i13n="cpos:11;pos:1">reportedly nixed plans</a> to put cameras in the Apple Watch, and Bloomberg suggests it’s opting instead to focus on delivering its own smart glasses this year. But it’s not hard to imagine that faster hardware could let the Apple Watch handle more Siri and AI-related tasks on its own. It’s already a fairly self-sufficient device, allowing you to ask basic Siri queries, run apps and listen to music without an iPhone — the cellular models are even more capable since they can take calls and send messages.Rumors also point to <a target="_blank" class="link" href="https://appleinsider.com/articles/25/10/12/airpods-pro-4-airpods-5-rumored-to-get-apple-intelligence-enhanced-cameras-in-2026#:~:text=AirPods%20Pro%204%2C%20AirPods%205,Intelligence%2Denhanced%20cameras%20in%202026&text=Apple's%20plans%20for%20the%20next,iPhones%20or%20Apple%20Vision%20Pro." data-i13n="cpos:12;pos:1">infrared cameras coming to the next AirPods and AirPod Pros</a>. Instead of taking photos, they could enable hand gestures and environmental awareness, which might be useful for Apple Intelligence down the line. The addition of <a target="_blank" class="link" href="https://www.engadget.com/audio/headphones/how-to-use-the-airpods-pro-3s-heart-rate-tracking-feature-141510469.html" data-i13n="cpos:13;pos:1">heart rate tracking in the AirPods Pro 3</a> shows that there are still new features Apple can bring to its buds, beyond listening to music.At best, an Apple AI pin could just be a simple way for someone to access Siri if they don’t want to wear an Apple Watch, plug in AirPods or have their iPhone within shouting distance. But at least those devices do other things beyond talking to Siri. The same is true for Meta’s Ray-Bans and future smart glasses. Even without accessing AI, they’ll still let you listen to music, take calls and, well, be glasses for those who need prescription frames.Given the vocal pushback against <a target="_blank" class="link" href="https://www.washingtonpost.com/technology/2025/08/27/gen-z-smart-glasses-digital-privacy-consent/" data-i13n="cpos:14;pos:1">Meta's Ray-Ban smart glasses</a>, which are also being <a target="_blank" class="link" href="https://www.androidcentral.com/wearables/you-can-no-longer-wear-xreal-or-metas-smart-glasses-in-public-on-this-cruise-ship" data-i13n="cpos:15;pos:1">banned on cruises</a>, clubs and other venues, I'm also not convinced many people would be eager to prominently display a surveillance device throughout the day. <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=fc65353f-45dc-4cd0-89be-ac2071680abb&siteId=us-engadget&pageId=1p-autolink&contentUuid=6a9c4806-23e0-4b7e-8496-a6afa6f6c483&featureId=text-link&merchantName=WIRED&linkText=Wired%E2%80%99s+Julian+Chokkattu+was+questioned&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy53aXJlZC5jb20vcmV2aWV3L2h1bWFuZS1haS1waW4vIiwiY29udGVudFV1aWQiOiI2YTljNDgwNi0yM2UwLTRiN2UtODQ5Ni1hNmFmYTZmNmM0ODMiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3LndpcmVkLmNvbS9yZXZpZXcvaHVtYW5lLWFpLXBpbi8ifQ&signature=AQAAAeECur4vIAclexvrByd8h2ampdv3Pn5Fj3822SEo2aBG&gcReferrer=https%3A%2F%2Fwww.wired.com%2Freview%2Fhumane-ai-pin%2F" data-i13n="elm:affiliate_link;sellerN:WIRED;elmt:;cpos:16;pos:1" data-original-link="https://www.wired.com/review/humane-ai-pin/">Wired’s Julian Chokkattu was questioned</a> about wearing a camera while he was testing the Humane AI Pin, and I’ve also had to explain to curious people why I was wearing Xreal’s smart glasses, which feature a prominent camera accessory.Sure, we're already living in a panopticon of smartphone cameras, but it's also obvious when someone is using their phone to capture photos and video. An AI pin just dangling off of your clothes is a constant threat, an unblinking eye. Even if Apple implements some sort of capture notification, <a target="_blank" class="link" href="https://www.bbc.com/news/articles/cx23ke7rm7go" data-i13n="cpos:17;pos:1">someone will always try to circumvent it</a>.While The Information notes Apple's AI pin may never actually see the light of day, I wouldn't be surprised if it does. This is the company that partnered with OpenAI just to make Siri appear slightly smarter with the debut of Apple Intelligence. And instead of building its own home-brewed AI models, it's <a target="_blank" class="link" href="https://www.engadget.com/ai/apples-siri-ai-will-be-powered-by-gemini-153636649.html" data-i13n="cpos:18;pos:1">banking on Google's Gemini to power Siri's big AI upgrade</a>, as well as its future foundation models. When it comes to AI, Apple will do almost anything to avoid being seen as a straggler (and to <a target="_blank" class="link" href="https://www.axios.com/2025/07/10/ai-apple-stock-market" data-i13n="cpos:19;pos:1">avoid even more stock declines</a>). It’s genuinely strange that Apple, the company that let Samsung and Google get a multi-year head start on foldable smartphones and hasn't yet jumped into the world of smart rings, could fast-track an AI pin for 2027. It’s yet another example of how the AI hype cycle has warped priorities throughout the tech industry. But at least Apple’s fortunes don’t depend on standalone AI hardware as much as OpenAI.  This article originally appeared on Engadget at https://www.engadget.com/ai/an-ai-pin-is-beneath-apple-182744647.html?src=rss

Jan 22, 2026

Korg's new experimental synthesizer combines acoustic sounds with electronic control

Korg just officially unveiled the <a data-i13n="cpos:1;pos:1" href="https://www.korg.com/us/products/dj/phase8/"><ins>Phase8 synthesizer</ins></a>, after years of tinkering by a team led by legendary engineer Tatsuya Takahashi. This is a highly experimental synthesizer, but it's not a concept unit. It'll be available for purchase in April. The Phase8 uses a new form of "acoustic synthesis" that combines acoustic sound generation with electronic control. Takahashi <a data-i13n="cpos:2;pos:1" href="https://www.musicradar.com/music-tech/its-beyond-analogue-vs-digital-its-even-beyond-electronics-korg-releases-experimental-phase8-eight-voice-acoustic-synthesizer"><ins>says the synthesizer</ins></a> is "beyond analog vs. digital" and "beyond electronics" altogether. It features chromatically tuned steel resonators, which creates an acoustic sound similar to that of a kalimba. These signals can be manipulated via onboard effects and sequenced like a traditional synthesizer. Here's a video of the synth in action. <div id="0bed0a4d53bc4780853b6ab8852c4d3e"><iframe width="560" height="315" src="https://www.youtube.com/embed/_w9PgAj4Llc?si=wwa9U6PMb4pV3Jxq" title="YouTube video player" frameborder="0" allowfullscreen></iframe></div> Players can mess with the resonators just like they would a guitar string or a piano key. Korg recommends that users physically touch, pluck, strum and tap the resonators to create new textures and tones. The company even suggests that people actually put objects on top of each resonator for additional experimentation. There's a slider that boots or dampens the acoustic response. As for the electronics, the Phase8 boasts dedicated envelopes and velocity control for each resonator and there's a polymetric step sequencer. The sequences can be stored and recalled via eight memory slots. All controls can be automated across a sequence. There are a handful of modulation effects, including tremolo and pitch-shift. <div id="e6eed28a23e043409ec5945a0fd632b8"><iframe width="560" height="315" src="https://www.youtube.com/embed/Nov6UT19OMk?si=MzMZiL-hRHrJgVjd" title="YouTube video player" frameborder="0" allowfullscreen></iframe></div> It features modern connectivity options, including MIDI in/out via 3.5mm, USB MIDI, a standard ¼-inch audio out and a headphone jack, among other options. Preorders are open right now, but hold on to your jaw. This thing costs $1,150.This article originally appeared on Engadget at https://www.engadget.com/audio/korgs-new-experimental-synthesizer-combines-acoustic-sounds-with-electronic-control-180809332.html?src=rss

Jan 22, 2026

Grok generated an estimated 3 million sexualized images — including 23,000 of children — over 11 days

We already knew xAI's Grok was barraging X with <a target="_blank" class="link" href="https://www.engadget.com/ai/elon-musks-grok-ai-posted-csam-image-following-safeguard-lapses-140521454.html" data-i13n="slk:nonconsensual sexual images of real people;cpos:1;pos:1">nonconsensual sexual images of real people</a>. But now there are some numbers to put things in perspective. Over an 11-day period, Grok generated an estimated 3 million sexualized images — including an estimated 23,000 of children.Put another way, Grok generated an estimated 190 sexualized images per minute during that 11-day period. Among those, it made a sexualized image of children once every 41 seconds.On Thursday, the Center for Countering Digital Hate (CCDH) <a target="_blank" class="link" href="https://counterhate.com/research/grok-floods-x-with-sexualized-images/" data-i13n="slk:published its findings;cpos:2;pos:1">published its findings</a>. The British nonprofit based its findings on a random sample of 20,000 Grok images from December 29 to January 9. The CCDH then extrapolated a broader estimate based on the 4.6 million images Grok generated during that period.The research defined sexualized images as those with "photorealistic depictions of a person in sexual positions, angles, or situations; a person in underwear, swimwear or similarly revealing clothing; or imagery depicting sexual fluids." The CCDH didn't take image prompts into account, so the estimate doesn't differentiate between nonconsensual sexualized versions of real photos and those generated exclusively from a text prompt.The CCDH used an AI tool to identify the proportion of the sampled images that were sexualized. That may warrant some degree of caution in the findings. However, I'm told that many third-party analytics services for X have reliable data because they use the platform's API.On January 9, xAI restricted Grok's ability to edit existing images to paid users. (That didn't solve the problem; it merely turned it into a premium feature.) Five days later, X <a target="_blank" class="link" href="https://www.engadget.com/ai/x-says-grok-will-no-longer-edit-images-of-real-people-into-bikinis-231430257.html" data-i13n="slk:restricted Grok's ability to digitally undress real people;cpos:3;pos:1">restricted Grok's ability to digitally undress real people</a>.<figure><img src="https://d29szjachogqwa.cloudfront.net/images/2026-01/440c8f59-6000-4b81-a2d8-c404e6f5545f" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/2026-01/440c8f59-6000-4b81-a2d8-c404e6f5545f" style="height:3852px;width:5778px;" alt="WASHINGTON, DC - JUNE 23: Google CEO Sundar Pichai (L) and Apple CEO Tim Cook (R) listen as U.S. President Joe Biden speaks during a roundtable with American and Indian business leaders in the East Room of the White House on June 23, 2023 in Washington, DC. Biden and Indian Prime Minister Narendra Modi held the meeting to meet with a range of leaders from the tech and business worlds and to discuss topics including innovation and AI. (Photo by Anna Moneymaker/Getty Images)" data-uuid="bcce1944-dd0f-3a8d-b1f7-4a39a1e3f4f5"/><figcaption>WASHINGTON, DC - JUNE 23: Google CEO Sundar Pichai (L) and Apple CEO Tim Cook (R) listen as U.S. President Joe Biden speaks during a roundtable with American and Indian business leaders in the East Room of the White House on June 23, 2023 in Washington, DC. Biden and Indian Prime Minister Narendra Modi held the meeting to meet with a range of leaders from the tech and business worlds and to discuss topics including innovation and AI. (Photo by Anna Moneymaker/Getty Images)</figcaption><div class="photo-credit">Anna Moneymaker via Getty Images</div></figure>But that restriction only applied to X; the standalone Grok app <a target="_blank" class="link" href="https://www.washingtonpost.com/technology/2026/01/15/grok-ai-image-generator-sexualized/" data-i13n="slk:reportedly;cpos:4;pos:1">reportedly</a> continues to generate these images. Since Apple and Google host the apps — which their policies explicitly prohibit — you might expect them to remove them from their stores. Well, in that case, you'd be wrong.So far, Tim Cook's Apple and Sundar Pichai's Google haven’t removed Grok from their stores — unlike similar “nudifying” apps from other developers. The companies also didn’t take any action on X while it was producing the images. That’s despite 28 women’s groups (and other progressive advocacy nonprofits) <a target="_blank" class="link" href="https://www.engadget.com/big-tech/28-advocacy-groups-call-on-apple-and-google-to-ban-grok-x-over-nonconsensual-deepfakes-215048460.html" data-i13n="slk:publishing an open letter calling on the companies to act;cpos:5;pos:1">publishing an open letter calling on the companies to act</a>.The companies haven't replied to multiple requests for comment from Engadget. To my knowledge, they haven't acknowledged the issue publicly in any format, nor have they responded to questions from other media outlets.<figure><img src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/grok_stores.jpg" data-crop-orig-src="https://d29szjachogqwa.cloudfront.net/images/user-uploaded/grok_stores.jpg" style="height:1136px;width:2019px;" alt="Grok - App Store and Play Store listings" data-uuid="9e99063c-1dfd-4b6c-b911-9d8571994004"/><figcaption>Grok - App Store and Play Store listings</figcaption><div class="photo-credit">Apple / Google</div></figure>The research’s findings on sexualized images included numerous outputs of people wearing transparent bikinis or micro-bikinis. The CCDH referred to one of a "uniformed healthcare worker with white fluids visible between her spread legs." Others included women wearing only dental floss, Saran Wrap or transparent tape. One depicted Ebba Busch, Sweden's Deputy Prime Minister, "wearing a bikini with white fluid on her head."Other public figures were part of that group. They include Selena Gomez, Taylor Swift, Billie Eilish, Ariana Grande, Ice Spice, Nicki Minaj, Christina Hendricks, Millie Bobby Brown and Kamala Harris.Examples of children include someone using Grok to edit a child's "before-school selfie" into an image of her in a bikini. Another image depicted "six young girls wearing micro bikinis." The CCDH said that, as of January 15, both of these posts were still live on X.In total, 29 percent of the sexualized images of children identified in the sample were still accessible on X as of January 15. The research found that even after posts were removed, the images remained accessible via their direct URLs.You can <a target="_blank" class="link" href="https://counterhate.com/research/grok-floods-x-with-sexualized-images/" data-i13n="slk:read the CCDH's report;cpos:6;pos:1">read the CCDH's report</a> for more details on the results and methodology. We’ll update this story if we receive a reply from Apple or Google.This article originally appeared on Engadget at https://www.engadget.com/ai/grok-generated-an-estimated-3-million-sexualized-images--including-23000-of-children--over-11-days-175053250.html?src=rss

Jan 22, 2026

How to watch the 2026 Super Bowl: TV channel, where to stream and more

We might still be in the midst of the playoffs, but that doesn't mean we can't plan ahead for <a data-i13n="cpos:1;pos:1" href="https://sports.yahoo.com/nfl/topic/super-bowl/">Super Bowl LX</a>. For those of you who just can't with Roman numerals, thats Super Bowl 60, and it's taking place this year at Levi's Stadium in Santa Clara, CA, on February 8, starting at 6:30 p.m. ET.  Like all other Sunday Night Football games this season, the championship game will be broadcast on Super Bowl Sunday on NBC, and will stream live on Peacock. And it's not just the game that we're excited for, either. This year's halftime performer is singer and rapper Bad Bunny, and there will be pre-game performances by Charlie Puth, Brandi Carlile, and Coco Jones, it's truly an incredible lineup of talent. So while we wait to find out which teams will be competing in this year's game, the least we can do is let you know everything you need to know to tune in to Super Bowl LX when it airs on Feb. 8. <h2 id="jump-link-how-to-watch-the-los-angeles-chargers-vs-new-england-patriots-this-sunday">How to watch Super Bowl LX</h2> <core-commerce id="6385b01fd93a46e1bead9be2aa70df09" data-type="product-list" data-original-url="https://www.peacocktv.com/?"></core-commerce> <core-commerce id="35e62e3779834329873e9458dc04c1ac" data-type="product-list" data-original-url="https://www.directv.com/stream/"></core-commerce> Date: Sunday, Feb. 8, 2026 Time: 6:30 p.m. ET TV channel: NBC, Telemundo Streaming: Peacock, DirecTV, NFL+ and more <h2 id="jump-link-los-angeles-chargers-vs-new-england-patriots-game-time">2026 Super Bowl game time</h2> The 2026 Super Bowl is set to begin at 6:30 p.m. ET/3:30 p.m. PT on Feb. 8, 2026. <h2 id="jump-link-los-angeles-chargers-vs-new-england-patriots-game-channel">2026 Super Bowl game channel</h2> The 2026 Super Bowl will air on NBC, with a Spanish-language broadcast available on Telemundo.  <h2 id="jump-link-where-is-the-2026-super-bowl-being-played">Where is the 2026 Super Bowl being played?</h2> The 2026 Super Bowl will be held at Levi's Stadium in Santa Clara, CA, home of the San Francisco 49ers. <h2 id="jump-link-what-teams-are-playing-in-the-2026-super-bowl">What teams are playing in the 2026 Super Bowl?</h2> The teams for the 2026 Super Bowl will be determined after the AFC and NFC Championship games are played on Sunday, Jan. 25. You can keep tabs on the post-season <a data-i13n="cpos:2;pos:1" href="https://sports.yahoo.com/nfl/playoff-bracket/">playoff bracket here</a>. <h2 id="jump-link-how-to-watch-the-los-angeles-chargers-vs-new-england-patriots-game-without-cable">How to watch the 2026 Super Bowl without cable</h2> You can stream NBC and Telemundo on platforms like DirecTV and Hulu + Live TV, both of which are among Engadget's choices for <a data-i13n="cpos:3;pos:1" href="https://www.engadget.com/entertainment/streaming/best-live-tv-streaming-service-133000410.html">best streaming services for live TV</a>. (Note that <a data-i13n="cpos:4;pos:1" href="https://www.engadget.com/entertainment/streaming/missing-nbc-on-fubo-heres-how-to-watch-sunday-night-football-figure-skating-and-more-this-week-014052346.html">Fubo and NBC are currently in the midst of a contract dispute</a> and NBC channels are not available on the platform.) The game will also be streaming on Peacock and on NFL+, though with an NFL+ subscription, you're limited to watching the game on mobile devices.  <core-commerce id="d9b771e8a71a4bb0859d8c7f2cb6a3f3" data-type="product-list" data-original-url="https://www.peacocktv.com/"></core-commerce> <core-commerce id="dab4c414b8fe40b29c2ec2d4ddd242e3" data-type="product-list" data-original-url="https://www.directv.com/stream/"></core-commerce> <core-commerce id="b3c0cc65f0674fd4976ab9e2010d2912" data-type="product-list" data-original-url="https://www.amazon.com/dp/B0F542J65B"></core-commerce> <h2 id="jump-link-who-is-performing-at-the-2026-super-bowl">Who is performing at the 2026 Super Bowl halftime show?</h2> Bad Bunny, who holds the title as the most-streamed artist in the world, will be headlining the 2026 Super Bowl halftime performance. You can expect that show to begin after the second quarter, likely between 8-8:30 p.m. ET. Singer Charlie Puth will also be at the game to perform the National Anthem, Brandi Carlile is scheduled to sing "America The Beautiful," and Coco Jones will perform "Lift Every Voice and Sing." <h2 id="jump-link-">Where to buy tickets to the 2026 Super Bowl: </h2> Tickets to the 2026 Super Bowl are available on third-party resale platforms like StubHub and Gametime.  <a class="athena-button rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=3daa87f2-c6d3-40c7-9637-c8aa7895a6c1&siteId=us-engadget&pageId=1p-autolink&contentUuid=f3d4e4b2-2273-433c-87e7-fbfb4424f573&featureId=shopnow-button&merchantName=StubHub&linkText=Find+tickets+on+Stubhub&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5zdHViaHViLmNvbS9uZmwtdGlja2V0cy9ncm91cGluZy8xMjEiLCJjb250ZW50VXVpZCI6ImYzZDRlNGIyLTIyNzMtNDMzYy04N2U3LWZiZmI0NDI0ZjU3MyIsIm9yaWdpbmFsVXJsIjoiaHR0cHM6Ly93d3cuc3R1Ymh1Yi5jb20vbmZsLXRpY2tldHMvZ3JvdXBpbmcvMTIxIn0&signature=AQAAAfA8_JCpm3vjKOGqh4pBwj3yqwOUfyGoJ9hKT74uuukh&gcReferrer=https%3A%2F%2Fwww.stubhub.com%2Fnfl-tickets%2Fgrouping%2F121" id="0f6237dc9cdf480bad1c5c2ce893a853" data-i13n="elm:affiliate_link;sellerN:StubHub;elmt:;sec:shop-now;cpos:1;pos:1" data-original-link="https://www.stubhub.com/nfl-tickets/grouping/121">Find tickets on Stubhub</a> <h2 id="jump-link-every-way-to-watch-nfl-games-this-season"></h2> <a class="athena-button rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=3b9fa87c-42fc-4c0d-9b00-70b0827f43ac&siteId=us-engadget&pageId=1p-autolink&contentUuid=f3d4e4b2-2273-433c-87e7-fbfb4424f573&featureId=shopnow-button&merchantName=Gametime&linkText=Find+tickets+on+Gametime&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL2dhbWV0aW1lLmNvL3N1cGVyLWJvd2wtdGlja2V0cy9wZXJmb3JtZXJzL25mbHN1cGVyYm93bCIsImNvbnRlbnRVdWlkIjoiZjNkNGU0YjItMjI3My00MzNjLTg3ZTctZmJmYjQ0MjRmNTczIiwib3JpZ2luYWxVcmwiOiJodHRwczovL2dhbWV0aW1lLmNvL3N1cGVyLWJvd2wtdGlja2V0cy9wZXJmb3JtZXJzL25mbHN1cGVyYm93bCJ9&signature=AQAAARDhhK4Cl3oKObRC0QYWtu4kKAM3n8_eswi90EkuE_UK&gcReferrer=https%3A%2F%2Fgametime.co%2Fsuper-bowl-tickets%2Fperformers%2Fnflsuperbowl" id="a5f40e134e584960b3e9358aa41e760c" data-i13n="elm:affiliate_link;sellerN:Gametime;elmt:;sec:shop-now;cpos:1;pos:1" data-original-link="https://gametime.co/super-bowl-tickets/performers/nflsuperbowl">Find tickets on Gametime</a> <h2 id="jump-link-more-ways-to-watch-super-bowl-lx">More ways to watch Super Bowl LX</h2> <core-commerce id="a829a9b525e4434fafb21c1bf795bbf1" data-type="product-list"></core-commerce> <core-commerce id="c93d42229a3d4dd184a88108a3d9b229" data-type="product-list"></core-commerce> <core-commerce id="da7f283b92ce420391d1432d5bedda05" data-type="product-list" data-original-url="https://id.nfl.com/select-subscription?"></core-commerce> This article originally appeared on Engadget at https://www.engadget.com/entertainment/streaming/how-to-watch-the-2026-super-bowl-tv-channel-where-to-stream-and-more-173222003.html?src=rss

Jan 22, 2026

GeForce Now adds support for flight controllers

Jan 22, 2026

The best VPN deals: Up to 87 percent off ProtonVPN, Surfshark, ExpressVPN, NordVPN and more

In a chaotic world, one thing you can count on is your own common-sense steps toward<a data-i13n="cpos:1;pos:1" href="https://www.engadget.com/cybersecurity/12-steps-you-can-take-right-now-to-be-safer-online-130008335.html"> better cybersecurity</a>. Most of the holiday deals are over by now, but plenty of services are still offering excellent prices. With access to a <a data-i13n="cpos:2;pos:1" href="https://www.engadget.com/cybersecurity/vpn/what-is-a-vpn-and-what-can-you-do-with-one-161549146.html">virtual private network (VPN)</a>, you can stream TV shows and events from all over the world, protect your information from hackers and thwart online trackers. We strongly recommend <a data-i13n="cpos:3;pos:1" href="https://www.engadget.com/cybersecurity/vpn/how-a-vpn-works-and-why-you-should-care-143000250.html">using a VPN</a>, but you might get stuck with a substandard app if you jump on the very first deal you see. You might also mistakenly end up paying more than you want to, as even otherwise respectable VPNs sometimes frame their prices in misleading ways, with advertised deals not always as available as they seem to be. Even so, there are some great bargains on the table. Many of the <a data-i13n="cpos:4;pos:1" href="https://www.engadget.com/cybersecurity/vpn/best-vpn-130004396.html">best VPNs</a> — including our top pick, Proton VPN — are still running deals that save you anywhere from 70 to 87 percent on annual subscriptions. Most of these discounts only apply if you sign up for a year or more, but if you divide the cost by the months of subscription, it's much cheaper over time. <h2 id="jump-link-best-vpn-deals">Best VPN deals</h2> <core-commerce id="96ac5765cf384f3bbb95c5bf40b9a658" data-type="product-list"></core-commerce> <a data-i13n="elm:affiliate_link;sellerN:ExpressVPN;elmt:;cpos:5;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=4255ebf1-6185-403e-9be2-91ab47ad1ac3&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=ExpressVPN&linkText=ExpressVPN+Basic+%E2%80%94+%2478.18+for+a+two-year+subscription+with+four+months+free+%2878+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5leHByZXNzdnBuLmNvbS90b3Avc3BlY2lhbC1kZWFsL29mZmVyIiwiY29udGVudFV1aWQiOiIzODIwZjA3Zi0wYzVlLTRhOTYtYmFiYS1kNmQ5ZWEwNTEwYzMiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3LmV4cHJlc3N2cG4uY29tL3RvcC9zcGVjaWFsLWRlYWwvb2ZmZXIifQ&signature=AQAAAdvyvaCYKEPEDRaiHSVqiB_O8xJWLFXjCV92TS8ZM1Yw&gcReferrer=https%3A%2F%2Fwww.expressvpn.com%2Ftop%2Fspecial-deal%2Foffer" class="rapid-with-clickid" data-original-link="https://www.expressvpn.com/top/special-deal/offer">ExpressVPN Basic — $78.18 for a two-year subscription with four months free (78 percent off):</a> This is one of the best VPNs, especially for new users, who will find its apps and website headache-free on all platforms. In tests for my <a data-i13n="cpos:6;pos:1" href="https://www.engadget.com/vpn-review-expressvpn-2023-gaming-streaming-160052492.html">ExpressVPN review</a>, it dropped my download speeds by less than 7 percent and successfully changed my virtual location 14 out of 15 times. In short, it's an all-around excellent service that only suffers from being a little overpriced — which is why I'm so excited whenever I find it offering a decent deal. This discount, which gets you 28 months of ExpressVPN service, represents a 78 percent savings. Be aware, though, that it'll renew at the $99.95 per year price. <a data-i13n="elm:affiliate_link;sellerN:ExpressVPN;elmt:;cpos:7;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=4255ebf1-6185-403e-9be2-91ab47ad1ac3&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=ExpressVPN&linkText=ExpressVPN+Advanced+%E2%80%94+%24100.58+for+a+two-year+subscription+with+four+months+free+%2874+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5leHByZXNzdnBuLmNvbS9vcmRlci8iLCJjb250ZW50VXVpZCI6IjM4MjBmMDdmLTBjNWUtNGE5Ni1iYWJhLWQ2ZDllYTA1MTBjMyIsIm9yaWdpbmFsVXJsIjoiaHR0cHM6Ly93d3cuZXhwcmVzc3Zwbi5jb20vb3JkZXIvIn0&signature=AQAAAYJJJiCVpBWMd-u8PFtNT1KgZesvJI5NwAjxhMUGyhNp&gcReferrer=https%3A%2F%2Fwww.expressvpn.com%2Forder%2F" class="rapid-with-clickid" data-original-link="https://www.expressvpn.com/order/">ExpressVPN Advanced — $100.58 for a two-year subscription with four months free (74 percent off):</a> ExpressVPN recently <a data-i13n="cpos:8;pos:1" href="https://www.engadget.com/cybersecurity/vpn/expressvpn-switches-to-multi-tiered-pricing-with-more-feature-options-130016633.html">split its pricing into multiple tiers</a>, but they all still come with similar discounts for going long. In addition to top-tier VPN service, advanced users get two additional simultaneous connections (for a total of 12), the ExpressVPN Keys password manager, advanced ad and tracker blocking, ID protection features and a 50 percent discount on an AirCove router. As above, note that it renews at $119.95 annually. <a data-i13n="elm:affiliate_link;sellerN:NordVPN;elmt:;cpos:9;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=c52a3d27-2d9a-44d8-8cf2-6f6387b122a0&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=NordVPN&linkText=NordVPN+Basic+%E2%80%94+%2481.36+for+a+two-year+subscription+%2870+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL25vcmR2cG4uY29tL3NwZWNpYWwvIiwiY29udGVudFV1aWQiOiIzODIwZjA3Zi0wYzVlLTRhOTYtYmFiYS1kNmQ5ZWEwNTEwYzMiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vbm9yZHZwbi5jb20vc3BlY2lhbC8ifQ&signature=AQAAAdOT-QW4mYUKKaZHKPgEaTgjP68Gxzw0j_z_9YlV2rsv&gcReferrer=https%3A%2F%2Fnordvpn.com%2Fspecial%2F" class="rapid-with-clickid" data-original-link="https://nordvpn.com/special/">NordVPN Basic — $81.36 for a two-year subscription (70 percent off):</a> NordVPN gets the most important parts of a VPN right. It's fast, it doesn't leak any of your data and it's great at changing your virtual location. I noted in my <a data-i13n="cpos:10;pos:1" href="https://www.engadget.com/cybersecurity/vpn/nordvpn-review-2025-innovative-features-a-few-missteps-163000578.html">NordVPN review</a> that it always connects quickly and includes a support page that makes it easy to get live help. NordVPN includes a lot of cool features, like servers that instantly connect you to Tor. This deal gives you 70 percent off the two-year plan. <a data-i13n="elm:affiliate_link;sellerN:NordVPN;elmt:;cpos:11;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=c52a3d27-2d9a-44d8-8cf2-6f6387b122a0&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=NordVPN&linkText=NordVPN+Plus+%E2%80%94+%2493.36+for+a+two-year+subscription+%2874+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL25vcmR2cG4uY29tL3NwZWNpYWwvcHJpY2luZy8iLCJjb250ZW50VXVpZCI6IjM4MjBmMDdmLTBjNWUtNGE5Ni1iYWJhLWQ2ZDllYTA1MTBjMyIsIm9yaWdpbmFsVXJsIjoiaHR0cHM6Ly9ub3JkdnBuLmNvbS9zcGVjaWFsL3ByaWNpbmcvIn0&signature=AQAAAYmuUj1Ii9goZGcp0SPeezjUlzb3qrEJ4zxaiRP4qpN8&gcReferrer=https%3A%2F%2Fnordvpn.com%2Fspecial%2Fpricing%2F" class="rapid-with-clickid" data-original-link="https://nordvpn.com/special/pricing/">NordVPN Plus — $93.36 for a two-year subscription (74 percent off):</a> NordVPN has also taken 70 percent off its Plus subscription. For only a little more, you get a powerful ad and tracker blocker that can also catch malware downloads, plus access to the NordPass password manager. A Plus plan also adds a data breach scanner that checks the dark web for your sensitive information. <a data-i13n="elm:affiliate_link;sellerN:surfshark;elmt:;cpos:12;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=6119f7ca-bf13-40ba-b3f6-2a335a8890f6&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=surfshark&linkText=Surfshark+Starter+%E2%80%94+%2453.73+for+a+two-year+subscription+with+three+months+free+%2887+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3N1cmZzaGFyay5jb20vcHJpY2luZyIsImNvbnRlbnRVdWlkIjoiMzgyMGYwN2YtMGM1ZS00YTk2LWJhYmEtZDZkOWVhMDUxMGMzIiwib3JpZ2luYWxVcmwiOiJodHRwczovL3N1cmZzaGFyay5jb20vcHJpY2luZyJ9&signature=AQAAAdH2FAIDIVtPS_0cqGWMA7QRjWMeFVrXRjSMMEO0fqYM&gcReferrer=https%3A%2F%2Fsurfshark.com%2Fpricing" class="rapid-with-clickid" data-original-link="https://surfshark.com/pricing">Surfshark Starter — $53.73 for a two-year subscription with three months free (87 percent off):</a> This is the "basic" level of <a data-i13n="cpos:13;pos:1" href="https://www.engadget.com/cybersecurity/vpn/surfshark-vpn-review-a-fast-vpn-for-casual-users-170022675.html">Surfshark</a>, but it includes the entire VPN; everything on Surfshark One is an extra perk. With this subscription, you'll get some of the most envelope-pushing features in the VPN world right now. Surfshark can rotate your IP constantly to help you evade detection — it even lets you choose your own entry and exit nodes for a double-hop connection. That all comes with a near-invisible impact on download speeds. With this year-round deal, you can save 87 percent on 27 months of Surfshark. <a data-i13n="elm:affiliate_link;sellerN:surfshark;elmt:;cpos:14;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=6119f7ca-bf13-40ba-b3f6-2a335a8890f6&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=surfshark&linkText=Surfshark+One+%E2%80%94+%2467.23+for+a+two-year+subscription+with+three+months+free+%2887+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3N1cmZzaGFyay5jb20vcHJpY2luZyIsImNvbnRlbnRVdWlkIjoiMzgyMGYwN2YtMGM1ZS00YTk2LWJhYmEtZDZkOWVhMDUxMGMzIiwib3JpZ2luYWxVcmwiOiJodHRwczovL3N1cmZzaGFyay5jb20vcHJpY2luZyJ9&signature=AQAAAdH2FAIDIVtPS_0cqGWMA7QRjWMeFVrXRjSMMEO0fqYM&gcReferrer=https%3A%2F%2Fsurfshark.com%2Fpricing" class="rapid-with-clickid" data-original-link="https://surfshark.com/pricing">Surfshark One — $67.23 for a two-year subscription with three months free (87 percent off):</a> A VPN is great, but it's not enough to protect your data all on its own. Surfshark One adds several apps that boost your security beyond just VPN service, including Surfshark Antivirus (scans devices and downloads for malware) and Surfshark Alert (alerts you whenever your sensitive information shows up in a data breach), plus Surfshark Search and Alternative ID from the tier below. This extra-low deal gives you 88 percent off all those features. If you bump up to Surfshark One+, you'll also get data removal through Incogni, but the price jumps enough that it's not quite worthwhile in my eyes. <a data-i13n="elm:affiliate_link;sellerN:CyberGhost;elmt:;cpos:15;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=9fd99815-3196-46cf-9b6f-707caf84c282&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=CyberGhost&linkText=CyberGhost+%E2%80%94+%2456.94+for+a+two-year+subscription+with+four+months+free+%2884+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5jeWJlcmdob3N0dnBuLmNvbS9idXkvY3liZXJnaG9zdC12cG4tMyIsImNvbnRlbnRVdWlkIjoiMzgyMGYwN2YtMGM1ZS00YTk2LWJhYmEtZDZkOWVhMDUxMGMzIiwib3JpZ2luYWxVcmwiOiJodHRwczovL3d3dy5jeWJlcmdob3N0dnBuLmNvbS9idXkvY3liZXJnaG9zdC12cG4tMyJ9&signature=AQAAAVFDOemP8POGnLzpyPpCaI_la8hdkR0Ve_jFdJWf5NL4&gcReferrer=https%3A%2F%2Fwww.cyberghostvpn.com%2Fbuy%2Fcyberghost-vpn-3" class="rapid-with-clickid" data-original-link="https://www.cyberghostvpn.com/buy/cyberghost-vpn-3">CyberGhost — $56.94 for a two-year subscription with four months free (84 percent off):</a> CyberGhost has some of the best automation you'll see on any VPN. With its Smart Rules system, you can determine how its apps respond to different types of Wi-Fi networks, with exceptions for specific networks you know by name. Typically, you can set it to auto-connect, disconnect or send you a message asking what to do. CyberGhost's other best feature is its streaming servers — I've found both better video quality and more consistent unblocking when I use them on streaming sites. Currently, you can get 28 months of CyberGhost for 79 percent off the usual price, but it'll renew at $56.94 per year. <a data-i13n="elm:affiliate_link;sellerN:Hide.me;elmt:;cpos:16;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=9e251e36-74b6-419a-8a7b-de182315da42&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=Hide.me&linkText=hide.me+%E2%80%94+%2469.95+for+a+two-year+subscription+with+four+months+free+%2875+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL2hpZGUubWUvZW4vcHJpY2luZyIsImNvbnRlbnRVdWlkIjoiMzgyMGYwN2YtMGM1ZS00YTk2LWJhYmEtZDZkOWVhMDUxMGMzIiwib3JpZ2luYWxVcmwiOiJodHRwczovL2hpZGUubWUvZW4vcHJpY2luZyJ9&signature=AQAAAe0QPv6SUX-6hvSv7_eROeuRZA8bFPmcBZA109HcavzL&gcReferrer=https%3A%2F%2Fhide.me%2Fen%2Fpricing" class="rapid-with-clickid" data-original-link="https://hide.me/en/pricing">hide.me — $69.95 for a two-year subscription with four months free (75 percent off):</a> Hide.me is an excellent free VPN — in fact, it's my favorite on the market, even with <a data-i13n="cpos:17;pos:1" href="https://www.engadget.com/cybersecurity/vpn/heres-how-eventvpn-is-different-from-other-free-vpns-213014671.html">EventVPN</a> and the free version of Proton VPN as competition. If you do want to upgrade to its paid plan, though, the two-year subscription offers great savings. Hide.me works well as a no-frills beginner VPN, with apps and a server network it should frankly be charging more for. <a data-i13n="elm:affiliate_link;sellerN:private internet access;elmt:;cpos:18;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=8b0732cc-efc2-4542-849d-cadc641525e2&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=private+internet+access&linkText=Private+Internet+Access+%E2%80%94+%2479+for+a+three-year+subscription+with+four+months+free+%2883+percent+off%29%3A&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3d3dy5wcml2YXRlaW50ZXJuZXRhY2Nlc3MuY29tL2J1eS12cG4tb25saW5lP2NvdXBvbj1vZmZpY2lhbC1zaXRlIiwiY29udGVudFV1aWQiOiIzODIwZjA3Zi0wYzVlLTRhOTYtYmFiYS1kNmQ5ZWEwNTEwYzMiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vd3d3LnByaXZhdGVpbnRlcm5ldGFjY2Vzcy5jb20vYnV5LXZwbi1vbmxpbmU_Y291cG9uPW9mZmljaWFsLXNpdGUifQ&signature=AQAAATlqHa_vZEKuis4XsOMk7-D6bWa2kZz1DudT0Xrs_4bN&gcReferrer=https%3A%2F%2Fwww.privateinternetaccess.com%2Fbuy-vpn-online%3Fcoupon%3Dofficial-site" class="rapid-with-clickid" data-original-link="https://www.privateinternetaccess.com/buy-vpn-online?coupon=official-site">Private Internet Access — $79 for a three-year subscription with four months free (83 percent off):</a> With this deal, you can get 40 months of <a data-i13n="cpos:19;pos:1" href="https://www.engadget.com/cybersecurity/vpn/private-internet-access-vpn-review-both-more-and-less-than-a-budget-vpn-120033882.html">Private Internet Access (PIA)</a> for a little bit under $2 per month — an 83 percent discount on its monthly price. You can also get a shorter one-year subscription for $40 ($3.33 per month). Despite being so cheap, PIA has plenty of features, coming with its own DNS servers, a built-in ad blocker and automation powers to rival CyberGhost. However, internet speeds can fluctuate while you're connected. <h2 id="jump-link-what-makes-a-good-vpn-deal">What makes a good VPN deal</h2> Practically every VPN heavily discounts its long-term subscriptions year-round, with even sharper discounts around occasions like the holidays. The only noteworthy exception is Mullvad, the Costco hot dog of VPNs (that's a compliment, to be clear). When there's constantly a huge discount going on, it can be hard to tell when you're actually getting a good deal. The best way to squeeze out more savings is to look for seasonal deals, student discounts or exclusive sales like <a data-i13n="elm:affiliate_link;sellerN:ProtonVPN;elmt:;cpos:20;pos:1" href="https://shopping.yahoo.com/rdlw?merchantId=06a20697-4135-4e83-a421-f326eee937ed&siteId=us-engadget&pageId=1p-autolink&contentUuid=3820f07f-0c5e-4a96-baba-d6d9ea0510c3&featureId=text-link&merchantName=ProtonVPN&linkText=Proton+VPN%27s+coupon+for+Engadget+readers&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL3Byb3RvbnZwbi5jb20vbC92cG4taG9tZS1wbGFucy1vZmZlci02NiIsImNvbnRlbnRVdWlkIjoiMzgyMGYwN2YtMGM1ZS00YTk2LWJhYmEtZDZkOWVhMDUxMGMzIiwib3JpZ2luYWxVcmwiOiJodHRwczovL3Byb3RvbnZwbi5jb20vbC92cG4taG9tZS1wbGFucy1vZmZlci02NiJ9&signature=AQAAAXblbAdg41Upy-ikI1NWGnvIj4qyrHoZiRe-XQqP16_n&gcReferrer=https%3A%2F%2Fprotonvpn.com%2Fl%2Fvpn-home-plans-offer-66" class="rapid-with-clickid" data-original-link="https://protonvpn.com/l/vpn-home-plans-offer-66">Proton VPN's coupon for Engadget readers</a>. One trick VPNs often use is to add extra months onto an introductory deal, pushing the average monthly price even lower. When it comes time to renew, you usually can't get these extra months again. You often can't even renew for the same basic period of time — for example, you may only be able to renew a two-year subscription for one year. If you're planning to hold onto a VPN indefinitely, check the fine print to see how much it will cost per month after the first renewal, and ensure that fits into your budget. Follow <a data-i13n="cpos:21;pos:1" href="https://twitter.com/EngadgetDeals">@EngadgetDeals</a> on X for the latest <a data-i13n="cpos:22;pos:1" href="https://www.engadget.com/deals/">tech deals</a> and <a data-i13n="cpos:23;pos:1" href="https://www.engadget.com/best-tech/">buying advice</a>.This article originally appeared on Engadget at https://www.engadget.com/deals/the-best-vpn-deals-up-to-88-percent-off-protonvpn-surfshark-expressvpn-nordvpn-and-more-120056445.html?src=rss

Jan 22, 2026

AMD's Ryzen 7 9850X3D CPU will arrive on January 29 and cost $499

Jan 22, 2026

Netflix scores Oscar nominations for Frankenstein and KPop Demon Hunters

The nominees for the 2026 Academy Awards <a data-i13n="cpos:1;pos:1" href="https://www.oscars.org/oscars/ceremonies/2026">were just announced</a> and Ryan Coogler's vampire-adjacent period drama Sinners received a record-breaking 16 nominations. That's the highest number of Oscar nominations for a single film in the history of the awards, followed by Titanic, La La Land and All About Eve with 14 noms each. Sinners grabbed a nomination in nearly every major category, including Best Picture, Best Director and Best Original Screenplay. The same goes for the performances, as Michael B. Jordan captured the nomination for Best Actor. Even Delroy Lindo got nominated for Best Supporting Actor, which is his <a data-i13n="cpos:2;pos:1" href="https://bsky.app/profile/discussingfilm.net/post/3mczcs45quc2r">very first Oscar nomination</a> after 50 years in the business. <div id="eb1639135cbb46e68139c7b39eacfa78"><iframe width="560" height="315" src="https://www.youtube.com/embed/ZaLpXqSoTlE?si=P9-GV3mzD7zgvedb" title="YouTube video player" frameborder="0" allowfullscreen></iframe></div> Paul Thomas Anderson's One Battle After Another walked away with 13 nominations and Yorgos Lanthimos' Bugonia scored four. Apple's F1 actually got nominated for Best Picture, which is unusual for a popcorn-style blockbuster. Netflix scored Best Picture noms for both Frankenstein and Train Dreams. Netflix also got its flowers for the international animated sensation KPop Demon Hunters. It got nominated for Best Animated Feature and, of course, Best Original Song. <div id="3ab711337c28452fb200281d3261c3e6"><iframe width="560" height="315" src="https://www.youtube.com/embed/yebNIHKAC4A?si=yFJG4J9VDhyJF68R" title="YouTube video player" frameborder="0" allowfullscreen></iframe></div> The 98th annual Academy Awards will take place on March 15, with Conan O'Brien returning to host. It'll be broadcast live on ABC. The awards ceremony <a data-i13n="cpos:3;pos:1" href="https://www.engadget.com/entertainment/tv-movies/the-oscars-will-exclusively-air-on-youtube-starting-in-2029-181545531.html">will be moving to YouTube</a> in 2029.This article originally appeared on Engadget at https://www.engadget.com/entertainment/tv-movies/netflix-scores-oscar-nominations-for-frankenstein-and-kpop-demon-hunters-163521141.html?src=rss

Jan 22, 2026

Waymo is now available in Miami

Jan 22, 2026

Xbox Developer Direct 2026: How to watch and what we expect to see from Fable, Forza Horizon 6 and Beast of Reincarnation

<div><div style="left:0;width:100%;height:0;position:relative;padding-bottom:56.25%;"><iframe src="https://www.youtube.com/embed/hXn6bvU1Kh4?rel=0" style="top:0;left:0;width:100%;height:100%;position:absolute;border:0;" allowfullscreen scrolling="no" data-embed-domain="youtube.com" data-provider-name="youtube"></iframe></div></div>Xbox is set to hold the fourth installment of its now-annual <a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/the-next-xbox-developer-direct-showcase-is-on-january-22-154444166.html" data-i13n="cpos:1;pos:1">Developer Direct showcase</a> on January 22. You’ll be able to watch the 2026 edition at 1PM ET on <a target="_blank" class="link" href="https://www.youtube.com/watch?v=hXn6bvU1Kh4" data-i13n="cpos:2;pos:1">YouTube</a> (including versions with <a target="_blank" class="link" href="https://www.youtube.com/live/AZZFG1WcJGA" data-i13n="cpos:3;pos:1">audio descriptions</a> and <a target="_blank" class="link" href="https://www.youtube.com/live/4BuRJESQZ9o" data-i13n="cpos:4;pos:1">ASL</a>), <a target="_blank" class="link" href="https://twitch.tv/Xbox" data-i13n="cpos:5;pos:1">Twitch</a>, <a target="_blank" class="link rapid-with-clickid" href="https://shopping.yahoo.com/rdlw?merchantId=58a137b8-4748-4ad2-b1db-003c79a332f5&siteId=us-engadget&pageId=1p-autolink&contentUuid=7820cad4-1943-4cb8-9f91-f84cce3a65c9&featureId=text-link&merchantName=Facebook&linkText=Facebook&custData=eyJzb3VyY2VOYW1lIjoiV2ViLURlc2t0b3AtVmVyaXpvbiIsImxhbmRpbmdVcmwiOiJodHRwczovL2ZhY2Vib29rLmNvbS9YYm94IiwiY29udGVudFV1aWQiOiI3ODIwY2FkNC0xOTQzLTRjYjgtOWY5MS1mODRjY2UzYTY1YzkiLCJvcmlnaW5hbFVybCI6Imh0dHBzOi8vZmFjZWJvb2suY29tL1hib3gifQ&signature=AQAAAYNWCOterXnBUEDGv85QXsXwHIj-PMgRm2nTL2fTh66s&gcReferrer=https%3A%2F%2Ffacebook.com%2FXbox" data-i13n="elm:affiliate_link;sellerN:Facebook;elmt:;cpos:6;pos:1" data-original-link="https://facebook.com/Xbox">Facebook</a> and Steam. Xbox will stream the Developer Direct on its regional channels simultaneously, and it’ll be available on Bilibili on Friday. The YouTube stream might be the way to go, since that one will be available in 4K and 60 fps. Because we’re so nice, we’ve embedded that stream above, so all you have to do is click the play button when the time is right.Microsoft has confirmed that it will feature three titles at the Developer Showcase: <a target="_blank" class="link" href="https://www.engadget.com/gaming/xbox/xbox-has-delayed-its-fable-reboot-until-2026-173118802.html" data-i13n="cpos:7;pos:1">Fable</a> and <a target="_blank" class="link" href="https://www.engadget.com/gaming/forza-horizon-6-takes-the-arcade-racing-series-to-japan-in-2026-142253601.html" data-i13n="cpos:8;pos:1">Forza Horizon 6</a> from Playground Games, and <a target="_blank" class="link" href="https://www.engadget.com/gaming/beast-of-reincarnation-is-a-one-person-one-dog-rpg-launching-in-2026-192305237.html" data-i13n="cpos:9;pos:1">Beast of Reincarnation</a> from Game Freak, which is best known for its work on Pokémon games. We should get developer interviews, gameplay footage and (for at least one of those) a release date. There is a chance that there will be some surprises, of course. The three previous editions of Developer Direct all featured five games — the first one showcased <a target="_blank" class="link" href="https://www.engadget.com/hi-fi-rush-release-date-205659375.html" data-i13n="cpos:10;pos:1">Hi-Fi Rush</a>, which was released on the same day.This article originally appeared on Engadget at https://www.engadget.com/gaming/xbox/xbox-developer-direct-2026-how-to-watch-and-what-we-expect-to-see-from-fable-forza-horizon-6-and-beast-of-reincarnation-161000242.html?src=rss

Jan 22, 2026

AI Mode in Google search can now pull context from your other apps

After <a target="_blank" class="link" href="https://www.engadget.com/ai/gemini-can-now-pull-context-the-rest-of-your-google-apps-if-you-let-it-160039468.html" data-i13n="cpos:1;pos:1">adding Personal Intelligence</a> to Gemini as an opt-in experience, Google has <a target="_blank" class="link" href="https://blog.google/products-and-platforms/products/search/personal-intelligence-ai-mode-search/" data-i13n="cpos:2;pos:1">announced</a> that it’s also integrating the feature into AI Mode in Search. What Personal Intelligence does is pull information from your Google apps to tailor its responses based on your history and interests. For Search, in particular, you can allow Personal Intelligence to look for information in your Gmail accounts and Google Photos libraries.If you use AI Mode to shop for clothes with the new feature enabled, for instance, Google could recommend items or models from a brand you previously purchased from. If it sees plane tickets or other reservations in Gmail, Google could also recommend specific items based on your destination and the season if you’re clothes shopping for that trip.Personal Intelligence is powered by Google’s <a target="_blank" class="link" href="https://www.engadget.com/ai/googles-new-gemini-3-model-arrives-in-ai-mode-and-the-gemini-app-160054273.html" data-i13n="cpos:3;pos:1">Gemini 3</a> AI model. The company says it doesn’t train its models using information from your Gmail inbox or Google Photos library, but it does use your prompts and AI Mode’s responses. Google also warned that sometimes, the feature’s recommendations could feel inaccurate because it could not fully comprehend the context or could make incorrect connections between separate topics. At the moment, Personal Intelligence is an experimental feature that’s rolling out in Labs starting today. Google AI Pro and Ultra subscribers in the US, who use the service in English, will automatically have access to it and be able to connect AI Mode to Gmail and Google Photos. It will only be available to personal Google accounts, however, and not for Workspace accounts just yet. This article originally appeared on Engadget at https://www.engadget.com/ai/ai-mode-in-google-search-can-now-pull-context-from-your-other-apps-160000103.html?src=rss

LWN

Jan 24, 2026

GNU C Library 2.43 released

Jan 23, 2026

[$] Filesystem medley: EROFS, NTFS, and XFS

Jan 23, 2026

GNU Guix 1.5.0 released

Jan 23, 2026

Two new stable kernels for Friday

Jan 23, 2026

Security updates for Friday

Jan 22, 2026

[$] Linux Kernel Runtime Guard reaches its 1.0 release

Jan 22, 2026

30 years of ReactOS

Jan 22, 2026

Rust 1.93.0 released

Jan 22, 2026

Security updates for Thursday

Jan 22, 2026

[$] LWN.net Weekly Edition for January 22, 2026

Jan 21, 2026

[$] Cleanup on aisle fsconfig()

Jan 21, 2026

Pandas 3.0 released

Jan 21, 2026

[$] Responses to gpg.fail

Jan 21, 2026

Security updates for Wednesday

Jan 20, 2026

Ryabitsev: Tracking kernel development with korgalore

Cloudflare Blog

Jan 23, 2026

Route leak incident on January 22, 2026

Jan 19, 2026

How we mitigated a vulnerability in Cloudflare’s ACME validation logic

Jan 16, 2026

Astro is joining Cloudflare

Jan 15, 2026

Human Native is joining Cloudflare

Jan 14, 2026

What came first: the CNAME or the A record?

Jan 13, 2026

What we know about Iran’s Internet shutdown

Jan 6, 2026

A closer look at a BGP anomaly in Venezuela

Dec 22, 2025

How Workers powers our internal maintenance scheduling pipeline

Dec 19, 2025

Code Orange: Fail Small — our resilience plan following recent incidents

Dec 19, 2025

Innovating to address streaming abuse — and our latest transparency report

Dec 18, 2025

Announcing support for GROUP BY, SUM, and other aggregation queries in R2 SQL

Dec 15, 2025

The 2025 Cloudflare Radar Year in Review: The rise of AI, post-quantum, and record-breaking DDoS attacks

Dec 15, 2025

ChatGPT's rivals, Kwai's quiet rise: the top Internet services of 2025

Dec 11, 2025

React2Shell and related RSC vulnerabilities threat brief: early exploitation activity and threat actor techniques

Dec 9, 2025

Shifting left at enterprise scale: how we manage Cloudflare with Infrastructure as Code

Dec 8, 2025

Python Workers redux: fast cold starts, packages, and a uv-first workflow

Dec 5, 2025

Cloudflare outage on December 5, 2025

Dec 3, 2025

Cloudflare WAF proactively protects against React vulnerability

Dec 3, 2025

Cloudflare's 2025 Q3 DDoS threat report -- including Aisuru, the apex of botnets

Dec 1, 2025

Why Replicate is joining Cloudflare

NPR

Jan 25, 2026

Russian strikes knock out heat in freezing Kyiv as peace talks continue

Jan 24, 2026

Photos: Massive winter storm sweeps across the U.S.

Jan 24, 2026

Man shot dead by federal immigration officers in Minneapolis

Jan 24, 2026

A man is fatally shot by federal agents in Minneapolis

Jan 24, 2026

Trump threatens Canada with 100% tariffs over its new trade deal with China

Jan 24, 2026

Heavy snow and rainfall kill 61, injure 110 over 3 days in Afghanistan

Jan 24, 2026

China's top general under investigation in latest military purge

Jan 24, 2026

Opinion: Mark Carney's warning and its echoes from the past

Jan 24, 2026

Why 3 Catholic cardinals released a statement critical of Trump's foreign policy

Jan 24, 2026

As the winter storm rages, here's what to know in your state

The Onion

Jan 23, 2026

Entire French Populace Moved To Tears By Cartoon Of Robot Holding Flower

Jan 23, 2026

Trump Boys Put Nobel Peace Prize In Microwave

Jan 23, 2026

Katy Perry’s Friends Skeptical Of Alleged Powerful Boyfriend Who Lives In Canada

Jan 23, 2026

Artist Profile: A$AP Rocky

Jan 23, 2026

Diana Chin and Tim Galchot

Jan 23, 2026

Onus Of Giving Parents Grandchildren Placed On Least Fucked-Up Sibling

Jan 23, 2026

‘House Burping’ Gains Popularity In U.S.

Jan 22, 2026

Usha Vance Announces Pregnancy With Fourth Child

Jan 22, 2026

ICE Deems Being In Privacy Of Own Home Obstruction Of Justice

Jan 22, 2026

Democrats Condemn ICE For Murdering Without Proper Warrants

Jan 22, 2026

Trump To Europe: ‘Getting Greenland Was Mr. Dilbert’s Final Wish’

Jan 22, 2026

Woman Throws Self On Grenade By Answering Call From Dysregulated Friend

New Scientist

Jan 19, 2026

Crowdsourcing Wikipedia’s encyclopedia: Best ideas of the century

Jan 19, 2026

The totemic 1.5°C climate target: Best ideas of the century

Jan 19, 2026

We can block the spread of HIV: Best ideas of the century

Jan 21, 2026

Peter F. Hamilton's latest is an epic slice of sci-fi – with one flaw

Jan 23, 2026

Bone cancer therapy unexpectedly makes tumours less painful

Jan 23, 2026

Sea turtles may be more resilient to global warming than we thought

Jan 19, 2026

The one diet that’s good for everything: Best ideas of the century

Jan 19, 2026

The electrification of everything: Best ideas of the century

Jan 19, 2026

Our solar system is extremely weird: Best ideas of the century

Jan 19, 2026

Smartphones (yes, really): Best ideas of the century

Jan 23, 2026

Why singing, dancing and engaging with art is good for your health

Jan 19, 2026

The hidden power of epigenetics: Best ideas of the century

Jan 19, 2026

End-to-end encryption: Best ideas of the century

Jan 19, 2026

Revealing the epic story of ancient humans: Best ideas of the century

Jan 19, 2026

How fear drastically shapes ecosystems: Best ideas of the century

Jan 23, 2026

Hybrid megapests evolving in Brazil are a threat to crops worldwide

Jan 19, 2026

The one innovation that supercharged AI: Best ideas of the century

Jan 19, 2026

There’s no such thing as a normal brain: Best ideas of the century

Jan 19, 2026

Filming the universe’s biggest dramas: Best ideas of the century

Jan 19, 2026

A revolution in how we do chemistry: Best ideas of the century

Jan 19, 2026

Pinning extreme weather on climate change: Best ideas of the century

Jan 23, 2026

SpaceX’s Starlink dodged 300,000 satellite collisions in 2025

Jan 23, 2026

Why biological clocks get our 'true age' wrong – and how AI could help

Jan 21, 2026

New Scientist recommends Avatar: Fire and Ash – especially the whale

Jan 22, 2026

Our oral microbiome could hold the key to preventing obesity

Jan 23, 2026

Why did magic mushrooms evolve? We may finally have the answer

Jan 21, 2026

Stunning images reveal the rich biodiversity of remotest Tanzania

Jan 21, 2026

Can we battle the downsides of a rule-based world, asks a new book

Jan 22, 2026

Ancient bacterium discovery rewrites the origins of syphilis

Jan 22, 2026

Ancient giant kangaroos could have hopped despite their huge size

Jan 19, 2026

The invention of net zero: Best ideas of the century

Jan 19, 2026

The immense interconnectivity of the brain: Best ideas of the century

Jan 21, 2026

The internet feels super lonely right now. Here's why

Jan 22, 2026

Does limiting social media help teens? We'll finally get some evidence

Jan 21, 2026

Ape-like hominin Paranthropus was more adaptable than we thought

Jan 22, 2026

Strips of dried placenta help wounds heal with less scarring

Jan 21, 2026

We were wrong about being able to 'nudge' people to improve the world

Jan 22, 2026

Piercing crocodile close-up wins ecology photo competition

Jan 21, 2026

Natural ovulation the best option before an IVF frozen embryo transfer

Jan 21, 2026

Our earliest vertebrate ancestors may have had four eyes

Jan 21, 2026

Oldest known rock art is a 68,000-year-old hand stencil with claws

Jan 21, 2026

Bird retinas work without oxygen, and now scientists know how

Jan 20, 2026

The 3 best ways to tackle anxiety, according to a leading expert

Jan 21, 2026

Alex Garland’s The Bone Temple is brutal, brilliant - and mind-blowing

Jan 21, 2026

Octopuses prompt rethink of why animals evolve big brains

Jan 20, 2026

Chernobyl cooling systems have lost power but meltdown risk is low

Jan 21, 2026

Bubble feeding trick spreads through humpback whale social groups

Jan 20, 2026

Cross-training may be the key to a long life

Jan 20, 2026

Scientists investigate ‘dark oxygen’ in deep-sea mining zone

Jan 20, 2026

World is entering an era of 'water bankruptcy'

Jan 20, 2026

Satellites could use magnetic fields to avoid collisions

Jan 20, 2026

Mars once had a vast sea the size of the Arctic Ocean

Jan 20, 2026

Sunscreens made from ground-up wood reach an SPF of over 180

Jan 19, 2026

Why a tool-using cow could change how we see farm animals

Jan 19, 2026

The 5 worst ideas of the 21st century – and how they went wrong

Jan 19, 2026

Barnacle gloop could improve inflammatory bowel disease treatments

Jan 19, 2026

Should Europe boycott US tech over Greenland, and is it even possible?

Jan 19, 2026

Star appears to have vanished in a failed supernova

Jan 19, 2026

The most important second in the entire history of the universe

Jan 19, 2026

Giving astronauts tardigrade toughness will be harder than we hoped

Jan 19, 2026

Volcanoes had lower greenhouse gas emissions in Earth's past

Jan 14, 2026

AI is promising to revolutionise how we diagnose mental illness

Jan 16, 2026

Amateur mathematicians solve long-standing maths problems with AI

Jan 14, 2026

Psychiatry has finally found an objective way to spot mental illness

Jan 14, 2026

The science that will help you feel more fulfilled with your life

Jan 14, 2026

Three ways to become calmer this New Year that you haven't tried (yet)

Jan 13, 2026

The Pacific Islanders fighting to save their homes from catastrophe

Jan 14, 2026

A new book provides a toolkit to tackle anxiety. Can it really help?

Jan 17, 2026

First treaty to protect the high seas comes into force

Jan 12, 2026

Our elegant universe: rethinking nature’s deepest principle

Jan 16, 2026

Meat may play an unexpected role in helping people reach 100

Jan 16, 2026

How to finally get a grasp on quantum computing

Jan 16, 2026

Cancelling plans may be more socially acceptable than you think

Jan 16, 2026

Earliest ever supernova sheds light on the first stars

Jan 16, 2026

A leading use for quantum computers might not need them after all

Jan 15, 2026

Body fat supports your health in surprisingly complex ways

Jan 14, 2026

New Scientist recommends Why We Drink Too Much for Dry January

Jan 14, 2026

These striking photos are a window into the world of quantum physics

Jan 14, 2026

We're getting intimate with chatbots. A new book asks what this means

Jan 15, 2026

Lithium-ion batteries could last longer with chemical tweak

Jan 15, 2026

Distant 'little red dot' galaxies may contain baby black holes

Jan 15, 2026

Fossil may solve mystery of what one of the weirdest-ever animals ate

Jan 15, 2026

6 ways to help your children have a healthy relationship with food

Jan 14, 2026

China has applied to launch 200,000 satellites, but what are they for?

Jan 15, 2026

All major AI models risk encouraging dangerous science experiments

Jan 14, 2026

Why non-human culture should change how we see nature

Jan 16, 2026

Cheating just three times massively ups the chance of winning at chess

Jan 14, 2026

Woolly rhino genome recovered from meat in frozen wolf pup’s stomach

Jan 14, 2026

Sinking river deltas put millions at risk of flooding

Jan 14, 2026

T. rex took 40 years to become fully grown

Jan 14, 2026

We must completely change the way we build homes to stay below 2°C

Jan 14, 2026

Sooner-than-expected climate impacts could cost the world trillions

Jan 13, 2026

These small lifestyle tweaks can add a year to your life

Jan 13, 2026

The hunt for where the last Neanderthals lived

Jan 13, 2026

Greenland sharks survive for centuries with diseased hearts

Jan 12, 2026

We're about to simulate a human brain on a supercomputer

Jan 12, 2026

Pompeii’s public baths were unhygienic until the Romans took over

Jan 12, 2026

Quantum computers could help sharpen images of exoplanets

Jan 12, 2026

Is there an evolutionary reason for same-sex sexual behaviour?

Jan 12, 2026

Why it’s easy to be misunderstood when talking about probability

Phys

Jan 24, 2026

Mining genomes for cyst nematode resistance could enable better soybean harvests

Jan 24, 2026

Edison's 1879 bulb experiments may have unintentionally produced graphene

Jan 24, 2026

Why do onions and chips keep washing up on England's south coast? Here's the science

Jan 24, 2026

General ad campaign for climate action receives more public engagement than tailored approach, study finds

Jan 24, 2026

Why some messages are more convincing than others

Jan 24, 2026

New nanoparticle technology offers hope for hard-to-treat diseases

Jan 24, 2026

New study sheds light on the threat of 'marine darkwaves' to ocean life

Jan 24, 2026

Microplastics in the atmosphere: Higher emissions come from land areas than from the ocean, study finds

Jan 24, 2026

Bullying tied to higher suicide attempt risk for high school girls

Jan 24, 2026

An app that uses drone footage to track plastic litter on beaches is promoting voluntary cleanup in Ireland

Jan 24, 2026

Humanity's oldest known cave art has been discovered in Sulawesi

Jan 24, 2026

Why we believe what we click: How self-selected online information shapes beliefs more than passive exposure

Jan 24, 2026

How to get managers to say yes to flexible work arrangements

Jan 24, 2026

Takeout meals serve as both reward and comfort after work, study finds

Jan 24, 2026

Rats demonstrate ability to replay episodic memories in complex experimental settings

Jan 24, 2026

A new three-way single step rearrangement enables precise ring editing

Jan 24, 2026

A year after undermining Bredt's rule, scientists make cage-shaped, double-bonded molecules that defy expectations

Jan 24, 2026

DNA origami enables precise patterning of molecules on 2D semiconductors

Jan 24, 2026

Ancient Spanish trees reveal Mediterranean storms are intensifying

Jan 24, 2026

Why some Central Pacific El Niños die quickly while others linger for years

Jan 24, 2026

Using amino acids as fuels to make conductive graphene

Jan 24, 2026

Saturday Citations: A weird, extinct life form; cholesterol hacking; interspecies prosociality of whales

Jan 24, 2026

Icy storm threatens Americans with power outages, extreme cold

Jan 24, 2026

What weather apps sometimes miss about dangerous winter storm conditions

Jan 24, 2026

Radicalism, extremism, fundamentalism: International study finds numerous commonalities—and certain differences

Jan 24, 2026

Why does it feel like dangerous ocean creatures are ruining Australia's summer?

Jan 24, 2026

Q&A: The present and future of the ecosystem reflected in marine life

Jan 24, 2026

Q&A: Fairness and well-being in society

Jan 24, 2026

Western governors called to Washington as Colorado River impasse drags on

Jan 23, 2026

Reading how to be male: Boys' literature reflects the rise of aggressive masculinity

Quanta

Jan 23, 2026

Monster Neutrino Could Be a Messenger of Ancient Black Holes

Jan 21, 2026

How Animals Build a Sense of Direction

Jan 20, 2026

Two Twisty Shapes Resolve a Centuries-Old Topology Puzzle

Jan 16, 2026

Why There’s No Single Best Way To Store Information

Jan 14, 2026

String Theory Can Now Describe a Universe That Has Dark Energy

PC Gamer

Jan 25, 2026

Finally, a comprehensive overhaul for third-person movement animations in Fallout: New Vegas

Jan 25, 2026

Cancelled Prince of Persia: The Sands of Time remake claimed to have been 'close to the finish line'

Jan 24, 2026

Fallout lead Tim Cain worked 70+ hours a week for 2 years to make the classic RPG: 'I'm glad things have changed, that was unsustainable⁠—but it was also absolutely amazing'

Jan 24, 2026

UK judge rules that RuneScape gold is fair game when suing for theft, at least in this case where an ex-developer stole $700,000 worth of it

Jan 24, 2026

In defense of XCOM: Chimera Squad

Jan 24, 2026

WoW's rough 2.0 transmog overhaul is mostly fixed, and I'm left wondering why Blizzard didn't just take the easy win

Jan 24, 2026

Deadlock's new Street Brawl mode is like a better version of the infamous MOBA 'All Random, All Mid,' but I think it could stand to be even more chaotic

Jan 24, 2026

From throwing bear traps with telekinesis to piledriving a zombie into a tree, this new roguelike is a brilliantly accessible gateway into the world of turn-based tactics

Jan 24, 2026

Ubisoft shares plummet to lowest price in over 14 years amid its recently announced 'restructuring' centered on cancellations, layoffs, and generative AI

Jan 24, 2026

This horror game built from the bones of an abandoned FPS server and an accidental ARPG might be one of the strangest puzzlers I've ever played

Jan 24, 2026

Our diabolical PC gaming dungeons quiz is harder than a level 1 Molten Core solo run

Jan 24, 2026

Inscryption-inspired card battler The Killing Stone gets a rad new trailer showing off voice talent from Critical Role and Baldur's Gate 3, and it's out next month

Jan 24, 2026

After debuting in Nightdive's second remaster, one of the best mods for classic FPS Blood just released its concluding episode for everyone to play

Jan 24, 2026

This side-scrolling metroidvania inspired by Bloodborne just got completely free DLC, and the base game is 50% off until February

Jan 24, 2026

No More Room in Hell 2 will add new ways to play in its upcoming 1.0 launch, one of which 'plays like your favourite classic game modes from the late 2000s'

Jan 24, 2026

It's hard to fault Highguard for giving the internet the silent treatment—wouldn't you?

Jan 24, 2026

Everything you can get from Arknights: Endfield Twitch Drops and how to link accounts

Jan 23, 2026

Building an optimized factory is so important in Arknights: Endfield that someone made a simulator tool for sharing your best blueprints

Jan 23, 2026

Fallout 4 feels like a brand new game now that it's forcing me to take naps, drink plenty of water, and not save scum

Jan 23, 2026

One of Steam's most wishlisted games announces it's launching next week with a song so unhinged you've got to hear it for yourself

Jan 23, 2026

Mass Effect boss Michael Gamble is looking for a production director for the next game in the series: 'They’ll report to me and it’s gonna be awesome'

Jan 23, 2026

Creator of DMCA'd Cyberpunk VR mod takes ball and goes home, removing access to all his VR mods after receiving a second DMCA takedown from 505 Games and a warning from Patreon

Jan 23, 2026

'Came for the combat, stayed to be a factory worker': Anime gacha game players are discovering they were born to be factory builders in Arknights: Endfield

Jan 23, 2026

That team-based shooter with phantom horses that debuted at The Game Awards and then promptly disappeared is not actually dead: A showcase featuring 'gameplay deep dive, year one plans, and much more' is coming next week

Jan 23, 2026

I asked Garry Newman if he really was going to rename s&box as Garry's mod 2 after an overwhelming poll and he joked 'people are d*ckheads'

Jan 23, 2026

How to remaster Prince Of Persia: The Sands of Time yourself, since Ubisoft won't

Jan 23, 2026

More than 40 US Democrats call for a 'thorough investigation' of EA's Saudi-led buyout over risks of layoffs, studio closures, and 'coordinated anti-labor practices'

Jan 23, 2026

Microsoft Windows 365 goes down the day after Microsoft celebrates 'reimagining the PC as a cloud service that streams a Cloud PC'

Jan 23, 2026

Software engineer creates classic SimCity-style map of NYC—and argues that AI will be good for creatives, actually

Jan 23, 2026

Not even Microsoft is confident that Copilot is right for coding, as it's encouraging staff to try out Anthropic's AI model, according to one report

Jan 23, 2026

Minecraft's lead dev says Hytale is 'very polished' but he'd rather just play Minecraft, and Hytale boss says he's 'probably right'—for now

Jan 23, 2026

Spellcasters Chronicles is getting its second closed beta weekend with more spells, a ranked mode, and better optimisation: 'We want every player to be able to play our game'

Jan 23, 2026

Palworld developer Pocketpair requires job applicants prove their gamer credentials: 'If the number of Steam games played is 0, we basically pass on them'

Jan 23, 2026

It turns out, GOG players get more achievements than Steam users do

Jan 23, 2026

Panther Lake and 18A 'create a firm foundation on which to build the new Intel'

Jan 23, 2026

Embark reveals Arc Raiders' next four major monthly updates, which it calls the 'Escalation' phase, teasing new matchmaking, machines, and more, but our second new map is months away

Jan 23, 2026

Asus ROG Azoth 96 HE review

Jan 23, 2026

Master Chief voice actor says he never wants AI cloning his hunky tones: 'That's where we cross a line… I don't like it, and I would prefer that it not be done'

Jan 23, 2026

Fable is getting rid of the old games' horrible dog, who I hate

Jan 23, 2026

MSI Vector A18 HX review

Jan 23, 2026

Forza Horizon 6 finally returns to the wristband career mode the series has drifted so far away from: 'You'll start out as just a tourist in Japan'

Jan 23, 2026

Finally, a game other than War Thunder may have led to the release of classified military information

Jan 23, 2026

Making the invisible visible, this engineering artist's creation shows the incredible level to which modern devices fill our world with radio waves

Jan 23, 2026

The new Fable's morality system will live in 'shades of gray', which is a huge shift for a series where being too mean made literal devil horns grow out of your head

Jan 23, 2026

Valve warms up for the Steam Machine with a client update bristling with gamepad fixes and, oh, the betas UI is fancier now if you're into that

Jan 23, 2026

Deadlock just received a huge update, with 6 new heroes, overhauled patrons, and a new game mode—and it was all predicted with Domino's pizza

Jan 23, 2026

Thermalright Peerless Assassin 120 SE review

Jan 23, 2026

Highguard requires Secure Boot and Easy Anti-Cheat to run, leaving Linux and kernel-conscious gamers out in the cold

Jan 23, 2026

'We will f*ck up': The publisher of Against the Storm and Manor Lords is committed to keeping generative AI out, but it's easier said than done

Jan 23, 2026

Intel's in no hurry to get 14A all tooled up: 'It does not make sense to build out significant capacity there until we know that we have the customers that will accept that demand'

IEEE Spectrum

Jan 24, 2026

The Project G Stereo Was the Definition of Groovy

<img src="https://spectrum.ieee.org/media-library/color-photo-of-a-stereo-system-from-above-with-two-black-spheres-at-either-end-and-a-rectangular-wood-cabinet-containing-a-tune.jpg?id=62978356&width=1245&height=700&coordinates=0%2C62%2C0%2C63"/> Dizzy Gillespie was a fan. Frank Sinatra bought one for himself and gave them to his Rat Pack friends. Hugh Hefner acquired one for the Playboy Mansion. Clairtone Sound Corp.’s <a href="https://www.clairtone.ca/projectg/" rel="noopener noreferrer" target="_blank">Project G</a> high-fidelity stereo system, which debuted in 1964 at the National Furniture Show in Chicago, was squarely aimed at trendsetters. The intent was to make the sleek, modern stereo an object of desire.By the time the Project G was introduced, the Toronto-based Clairtone was already well respected for its beautiful, high-end stereos. “Everyone knew about Clairtone,” Peter Munk, president and cofounder of the company, boasted to a newspaper columnist. “The prime minister had one, and if the local truck driver didn’t have one, he wanted one.” Alas, with a price tag of CA $1,850—about the price of a small car—it’s unlikely that the local truck driver would have actually bought a Project G. But he could still dream.The design of the Project G seemed to come from a dream.“I want you to imagine that you are visitors from Mars and that you have never seen a Canadian living room, let alone a hi-fi set,” is how designer Hugh Spencer challenged Clairtone’s engineers when they first started working on the Project G. “What are the features that, regardless of design considerations, you would like to see incorporated in a new hi-fi set?” <img alt="Black and white photo of a young woman sitting on the floor in front of a stereo system and looking toward the floor." class="rm-shortcode" data-rm-shortcode-id="69c4984f1ac1e02a66c610fd59b2d838" data-rm-shortcode-name="rebelmouse-image" id="5848e" loading="lazy" src="https://spectrum.ieee.org/media-library/black-and-white-photo-of-a-young-woman-sitting-on-the-floor-in-front-of-a-stereo-system-and-looking-toward-the-floor.jpg?id=62979218&width=980"/> The film “I’ll Take Sweden” featured a Project G, shown here with co-star Tuesday Weld.Nina Munk/The Peter Munk EstateThe result was a stereo system like no other. Instead of speakers, the Project G had sound globes. Instead of the heavy cabinetry typical of <a href="https://spectrum.ieee.org/vintage-audio-repair" target="_self">1960s entertainment consoles</a>, it had sleek, angled rosewood panels balanced on an aluminum stand. At over 2 meters long, it was too big for the average living room but perfect for Hollywood movies—Dean Martin had one in his swinging Malibu bachelor pad in the 1965 film <a href="https://www.imdb.com/title/tt0059431/" target="_blank">Marriage on the Rocks</a>. According to the 1964 press release announcing the Project G, it was nothing less than “a new sculptured representation of modern sound.”The first-generation Project G had a high-end Elac Miracord 10H turntable, while later models used a Garrard Lab Series turntable. The transistorized chassis and control panel provided AM, FM, and FM-stereo reception. There was space for storing LPs or for an optional Ampex 1250 reel-to-reel tape recorder. The “G” in Project G stood for “globe.” The hermetically sealed 46-centimeter-diameter sound globes were made of spun aluminum and mounted at the ends of the cantilevered base; inside were Wharfedale speakers. The sound globes rotated 340 degrees to project a cone of sound and could be tuned to re-create the environment in which the music was originally recorded—a concert hall, cathedral, nightclub, or opera house. <img alt="Between 1965 and 1967, Clairtone sponsored the Miss Canada beauty pageant. Miss Canada 1963 was Diane Landry, seen here with a Project G2 at Clairtone\u2019s factory showroom in Rexdale, Ontario." class="rm-shortcode" data-rm-shortcode-id="4c9b6865b02166b3cada4ad97e1fe857" data-rm-shortcode-name="rebelmouse-image" id="66613" loading="lazy" src="https://spectrum.ieee.org/media-library/between-1965-and-1967-clairtone-sponsored-the-miss-canada-beauty-pageant-miss-canada-1963-was-diane-landry-seen-here-with-a-p.jpg?id=62979401&width=980"/> Diane Landry, winner of the 1963 Miss Canada beauty pageant, poses with a Project G2. Nina Munk/The Peter Munk EstateInitially, Clairtone intended to produce only a handful of the stereos. As one writer later put it, it was more like a concept car “intended to give Clairtone an aura of futuristic cool.” Eventually fewer than 500 were made. But the Project G still became an icon of mod ’60s Canadian design, winning a silver medal at the 13th Milan Triennale, the international design exhibition.And then it was over; the dream had ended. Eleven years after its founding, Clairtone collapsed, and Munk and cofounder David Gilmour lost control of the company.<h2>The birth of Clairtone Sound Corp.</h2>Clairtone’s Peter Munk lived a colorful life, with a nightmarish start and many fantastic and dreamlike parts too. He was born in 1927 in Budapest to a prosperous Jewish family. In the spring of 1944, Munk and 13 members of his family boarded a train with more than 1,600 Jews bound for the Bergen-Belsen concentration camp. They arrived, but after some weeks the train moved on, eventually reaching neutral Switzerland. It later emerged that the Nazis had extorted large sums of cash and valuables from the occupants in exchange for letting the train proceed.As a teenager in Switzerland, Munk was a self-described party animal. He enjoyed dancing and dating and going on long ski trips with friends. Schoolwork was not a top priority, and he didn’t have the grades to attend a Swiss university. His mother, an Auschwitz survivor, encouraged him to study in Canada, where he had an uncle.Before he could enroll, though, Munk blew his tuition money entertaining a young woman during a trip to New York. He then found work picking tobacco, earned enough for tuition, and graduated from the University of Toronto in 1952 with a degree in electrical engineering. <img alt="Color photo of two men in office attire. " class="rm-shortcode" data-rm-shortcode-id="c784cd77090f613e1dd2d1ebd7364730" data-rm-shortcode-name="rebelmouse-image" id="1c5ae" loading="lazy" src="https://spectrum.ieee.org/media-library/color-photo-of-two-men-in-office-attire.jpg?id=62980489&width=980"/> Clairtone cofounders Peter Munk [left] and David Gilmour envisioned the company as a luxury brand.Nina Munk/The Peter Munk EstateAt the age of 30, Munk was making custom hi-fi sets for wealthy clients when he and David Gilmour, who owned a small business importing Scandinavian goods, decided to join forces. Their idea was to create high-fidelity equipment with a contemporary Scandinavian design. Munk’s father-in-law, William Jay Gutterson, invested $3,000. Gilmour mortgaged his house. In 1958, Clairtone Sound Corp. was born. From the beginning, Munk and Gilmour sought a high-end clientele. They positioned Clairtone as a luxury brand, part of an elegant lifestyle. If you were the type of woman who listened to music while wearing pearls and a strapless gown and lounging on a shag rug, your music would be playing on a Clairtone. If you were a man who dressed smartly and owned an <a href="https://arnejacobsen.com/works/the-egg-2/" target="_blank">Arne Jacobsen Egg chair</a>, you would also be listening on a Clairtone. That was the modern lifestyle captured in the company’s advertisements. In 1958, Clairtone produced its first prototype: the monophonic 100-M, which had a long, low cabinet made from oiled teak, with a Dual 1004 turntable, a Granco tube chassis, and a pair of Coral speakers. It never went into production, but the next model, the stereophonic 100-S, won a Design Award from Canada’s National Industrial Design Council in 1959. By 1963, Clairtone was selling 25,000 units a year. <img alt="Black and white photo of a line of stereo components under assembly, with a man in a lab coat at one end and a man in a suit at the other. " class="rm-shortcode" data-rm-shortcode-id="c93ba9d388b5f08fc41195f503119686" data-rm-shortcode-name="rebelmouse-image" id="90768" loading="lazy" src="https://spectrum.ieee.org/media-library/black-and-white-photo-of-a-line-of-stereo-components-under-assembly-with-a-man-in-a-lab-coat-at-one-end-and-a-man-in-a-suit-at.jpg?id=62981917&width=980"/> Peter Munk visits the Project G assembly line in 1965. Nina Munk/The Peter Munk EstateDesign was always front and center at Clairtone, not just for the products but also for the typography, advertisements, and even the annual reports. Yet nothing in the early designs signaled the dramatic turn it would take with the Project G. That came about because of Hugh Spencer.Spencer was not an engineer, nor did he have experience designing consumer electronics. His day job was designing sets for the <a href="https://www.cbc.ca/" target="_blank">Canadian Broadcast Corp.</a> He consulted regularly with Clairtone on the company’s graphics and signage. The only stereo he ever designed for Clairtone was the Project G, which he first modeled as a wooden box with tennis balls stuck to the sides. From both design and quality perspectives, Clairtone was successful. But the company was almost always hemorrhaging cash. In 1966, with great fanfare and large government incentives, the company opened a state-of-the-art production facility in Nova Scotia. It was a mismatch. The local workforce didn’t have the necessary skills, and the surrounding infrastructure couldn’t handle the production. On 27 August 1967, Munk and Gilmour were forced out of Clairtone, which became the property of the government of Nova Scotia.Despite the demise of their first company (and the government inquiry that followed), Munk and Gilmour remained friends and went on to become serial entrepreneurs. Their next venture? A resort in Fiji, which became part of a large hotel chain in that country, Australia, and New Zealand. (Gilmour later founded <a href="https://www.fijiwater.com/" target="_blank">Fiji Water</a>.) Then Munk and Gilmour bought a gold mine and cofounded Barrick Gold (now <a href="https://www.barrick.com/English/home/default.aspx" target="_blank">Barrick Mining Corp.</a>, one of the largest gold mining operations in the world). Their businesses all had ups and downs, but both men became extremely wealthy and noted philanthropists.<h2>Preserving Canadian design</h2>As an example of iconic design, the Project G seems like an ideal specimen for museum collections. And in 1991, Frank Davies, one of the designers who worked for Clairtone, donated a Project G to the recently launched Design Exchange in Toronto. It would be the first object in the DX’s permanent collection, which sought to preserve examples of Canadian design. The museum quickly became Canada’s center for the promotion of design, hosting more than 50 programs each year to teach people about how design influences every aspect of our lives. In 2008, the museum opened The Art of Clairtone: The Making of a Design Icon, 1958–1971, an exhibition showcasing the company’s distinctive graphic design, industrial design, engineering, and photography. <img alt="Color photo of a modern stereo system in the foreground and a woman sitting in a modern arm chair in the back. " class="rm-shortcode" data-rm-shortcode-id="308374818a417d5be186d1b390ad6315" data-rm-shortcode-name="rebelmouse-image" id="24fd8" loading="lazy" src="https://spectrum.ieee.org/media-library/color-photo-of-a-modern-stereo-system-in-the-foreground-and-a-woman-sitting-in-a-modern-arm-chair-in-the-back.jpg?id=62988475&width=980"/> David Gilmour’s wife, Anna Gilmour, was the company’s first in-house model.Nina Munk/The Peter Munk EstateBut what happened to the DX itself is a reminder that any museum, however worthy, shouldn’t be taken for granted. In 2019, the DX abruptly closed its permanent collection, and curators were charged with deaccessioning its objects. Fortunately, the <a href="https://www.rom.on.ca/" target="_blank">Royal Ontario Museum</a>, <a href="https://carleton.ca/" target="_blank">Carleton</a> and <a href="https://www.yorku.ca/" target="_blank">York</a> Universities, and the <a href="https://www.archives.gov.on.ca/" rel="noopener noreferrer" target="_blank">Archives of Ontario</a>, among others, were able to accept the artifacts and companion archives. (The <a href="https://collections.rom.on.ca/objects/2476799/project-g-stereo" rel="noopener noreferrer" target="_blank">Project G</a> pictured at top is now at the Royal Ontario Museum.) Researchers at York and Carleton have been working to digitize and virtually reconstitute the DX collection, through the <a href="https://rs.lincsproject.ca/resource/search:xdx" rel="noopener noreferrer" target="_blank">xDX Project</a>. They’re using the Linked Infrastructure for Networked Cultural Scholarship (LINCS) to turn interlinked and contextualized data about the collection into a searchable database. It’s a worthy goal, even if it’s not quite the same as having all of the artifacts and supporting papers physically together in one place. I admit to feeling both pleased about this virtual workaround, and also a little sad that a unified collection that once spoke to the historical significance of Canadian design no longer exists.Part of a <a href="https://spectrum.ieee.org/collections/past-forward/" target="_self">continuing series</a> looking at historical artifacts that embrace the boundless potential of technology.An abridged version of this article appears in the February 2026 print issue as “The Project G Stereo Defined 1960s Cool.”<h3>References </h3> I first learned about Clairtone’s Project G from a panel on Canada’s design heritage organized by York University historian Jan Hadlaw at the 2025 annual meeting of the Society for the History of Technology.<a href="https://ninamunk.com/books-cpt/the-art-of-clairtone/" target="_blank">The Art of Clairtone: The Making of a Design Icon, 1958–1971</a> by Nina Munk (Peter Munk’s daughter) and Rachel Gotlieb (McClelland & Stewart, 2008) was the companion book to the <a href="https://hamblywoolley.com/work/the-design-exchange-clairtone-exhibit/" rel="noopener noreferrer" target="_blank">exhibition</a> of the same name hosted by the Design Exchange in Toronto. It was an invaluable resource for this column.Journalist Garth Hopkins’s Clairtone: The Rise and Fall of a Business Empire (McClelland & Stewart, 1978) includes many interviews with people associated with the company.<a href="https://www.sphinxproductions.com/films/clairtone" rel="noopener noreferrer" target="_blank">Clairtone</a> is a new documentary by Ron Mann that came out while I was writing this piece. I haven’t been able to view it yet, but I hope to do so soon.

Jan 23, 2026

Video Friday: Humans and Robots Team Up in Battlefield Triage

<img src="https://spectrum.ieee.org/media-library/four-legged-robot-with-camera-moving-across-grassy-terrain.png?id=63130606&width=1245&height=700&coordinates=0%2C64%2C0%2C64"/> Video Friday is your weekly selection of awesome robotics videos, collected by your friends at IEEE Spectrum robotics. We also post a weekly calendar of upcoming robotics events for the next few months. Please <a href="mailto:automaton@ieee.org?subject=Robotics%20event%20suggestion%20for%20Video%20Friday">send us your events</a> for inclusion.<h5><a href="https://2026.ieee-icra.org/">ICRA 2026</a>: 1–5 June 2026, VIENNA</h5>Enjoy today’s videos!<div class="horizontal-rule"></div><div style="page-break-after: always"> </div>One of my favorite parts of robotics is watching research collide with non-roboticists in the real (or real-ish) world.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/4Cc_kG-HeHA?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.darpa.mil/research/challenges/darpa-triage-challenge">DARPA</a> ]<div class="horizontal-rule"></div>Spot will <a data-linked-post="2674366356" href="https://spectrum.ieee.org/wildfire-drones" target="_blank">put out fires</a> for you. Eventually. If it feels like it.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/2wH8USs-2vo?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://asmedigitalcollection.asme.org/letterstransrobotics/article-abstract/1/3/031004/1229384/Development-of-an-Autonomous-Firefighting?redirectedFrom=fulltext">Mechatronic and Robotic Systems Laboratory</a> ]<div class="horizontal-rule"></div>All those robots rising out of their crates is not sinister at all.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/vYl-CPlnYp0?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.limxdynamics.com/en">LimX</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="tlq_oynn1sc">The Lynx M20 quadruped robot recently completed an extreme cold-weather field test in Yakeshi, Hulunbuir, operating reliably in temperatures as low as –30°C.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/TlQ_OYNn1Sc?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.deeprobotics.cn/en">DEEP Robotics</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="jq1vmt5l1zg">This is a teaser video for KIMLAB’s new teleoperation robot. For now, we invite you to enjoy the calm atmosphere, with students walking, gathering, and chatting across the UIUC Main Quad—along with its scenery and ambient sounds, without any technical details. More details will be shared soon. Enjoy the moment.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/jq1Vmt5L1Zg?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>The most incredible part of this video is that they have publicly available power in the middle of their quad.[ <a href="https://publish.illinois.edu/kimlab2020/">KIMLAB</a> ]<div class="horizontal-rule"></div>For the eleventy-billionth time: Just because you can do a task with a <a data-linked-post="2666662286" href="https://spectrum.ieee.org/humanoid-robots" target="_blank">humanoid robot</a> doesn’t mean you should do a task with a humanoid robot.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/xXiTvnsi4EI?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.ubtrobot.com/en/">UBTECH</a> ]<div class="horizontal-rule"></div>I am less interested in this autonomous urban delivery robot and more interested in whatever that docking station is at the beginning that loads the box into it.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/SMaDPiJMdjQ?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://unmanned.kaist.ac.kr/">KAIST</a> ]<div class="horizontal-rule"></div>Okay, so figuring out where <a data-linked-post="2650272183" href="https://spectrum.ieee.org/spot-is-boston-dynamics-nimble-new-quadruped-robot" target="_blank">Spot’s face</a> is just got a lot more complicated.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/de06_CLqrM0?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://bostondynamics.com/blog/a-new-perspective-for-facilities-inspection/">Boston Dynamics</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="da22c-akgzy">An undergraduate team at HKU’s Tam Wing Fan Innovation Wing developed CLIO, an embodied tour-guide robot, just in months. Built on LimX Dynamics TRON 1, it uses LLMs for tour planning, computer vision for visitor recognition, and a laser pointer/expressive display for engaging tours.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/DA22C-aKgZY?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://arxiv.org/html/2512.05389">CLIO</a> ]<div class="horizontal-rule"></div>The future of work is doing work so that robots can then do the same work, except less well.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/JfUFSR_xnqI?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://global.agilex.ai/">AgileX</a> ]<div class="horizontal-rule"></div>

Jan 22, 2026

Thinking of Joining IEEE’s Leadership Ranks?

<img src="https://spectrum.ieee.org/media-library/colorful-silhouettes-of-several-people-talking-to-each-other.jpg?id=63071207&width=1245&height=700&coordinates=0%2C469%2C0%2C469"/> Strong leadership is essential for IEEE to advance technology for humanity. The organization depends on the dedicated service of its volunteers to <a href="https://spectrum.ieee.org/ieee-presidents-note-dec-2025" target="_self">advance its mission</a>. Each year, the <a href="https://www.ieee.org/about/corporate/nominations" rel="noopener noreferrer" target="_blank">Nominations and Appointments (N&A) Committee</a> is responsible for recommending candidates to the Board of Directors and the IEEE Assembly for volunteer leadership positions, including <a href="https://spectrum.ieee.org/2027-ieee-president-elect-candidates" target="_self">president-elect</a>, corporate officers, committee chairs, and committee members. See below for the complete list. By nominating qualified, experienced, committed volunteers, you help ensure continuity, good governance, and thoughtful decision-making at the highest levels of the organization. We encourage nominators to take a deliberate approach and align nominations with each candidate’s demonstrated experience and the<a href="https://www.ieee.org/about/corporate/nominations/nac-position-descriptions.html" rel="noopener noreferrer" target="_blank"> specific qualifications</a> of the role.To nominate a person for a position, complete <a href="https://ieeenoms.secure-platform.com/a/organizations/main/home" rel="noopener noreferrer" target="_blank">this form</a>.The N&A Committee is currently seeking nominees for the following positions:<h3>2028 IEEE President-Elect (who will be elected in 2027 and will serve as President in 2029 )</h3><h3>2027 IEEE Corporate Officers</h3>• Secretary • Treasurer • Vice President, Educational Activities • Vice President, Publication Services and Products<h3>2027 IEEE Committees Chairs and Members</h3>• Audit • Awards Board • Collaboration and Engagement • Conduct Review • Election Oversight • Employee Benefits and Compensation • Ethics and Member Conduct • European Public Policy • Fellow • Fellow Nominations and Appointments • Governance • History • Humanitarian Technologies Board • Industry Engagement • Innovations (formerly New Initiatives) • Nominations and Appointments • Public Visibility • Tellers<h2>Deadlines for nominations</h2>15 March<ul><li>Vice President, Educational Activities</li><li>Vice President, Publication Services and Products</li><li>Committee Chairs</li></ul>15 June<ul><li>President-Elect</li><li>Secretary</li><li>Treasurer</li><li>Committee Members</li></ul><h2>Deadlines for self-nominations</h2>30 March<ul><li>Vice President, Educational Activities</li><li>Vice President, Publication Services and Products</li><li>Committee Chairs</li></ul>30 June<ul><li>President-Elect</li><li>Secretary</li><li>Treasurer</li><li>Committee Members</li></ul><h2>Who can nominate</h2>Anyone may submit a nomination. Self-nominations are encouraged. Nominators need not be IEEE members, but nominees must meet specific qualifications. An IEEE organizational unit may submit recommendations endorsed by its governing body or the body’s designee.A person may be nominated for more than one position, however nominators are encouraged to focus on positions that align closely with the candidate’s qualifications and experience. Nominators need not contact their nominees before submitting the form. The IEEE N&A committee will contact eligible nominees for the required documentation and for their interest and willingness to be considered for the position.<h2>How to nominate</h2>For information about the positions, including qualifications, estimates of the time required by each position during the term of office, and the nomination process check the <a href="https://www.ieee.org/about/corporate/nominations/" rel="noopener noreferrer" target="_blank">IEEE Nominations and Appointments Committee website</a>. To nominate a person for a position, complete <a href="https://ieeenoms.secure-platform.com/a/organizations/main/home" rel="noopener noreferrer" target="_blank">this form</a>.<h2>Nominating tips</h2>Make sure to check eligibility requirements on the <a href="https://www.ieee.org/about/corporate/nominations/nac-position-descriptions.html" rel="noopener noreferrer" target="_blank">N&A committee website</a> before submitting a nomination as those that do not meet the stated requirements will not be advanced.Volunteers with relevant prior experience in lower-level IEEE committees and units are recommended by the committee more often than volunteers without such experience.Individuals recommended for president-elect and corporate officer positions are more likely to be recommended if they possess a strong track record of leadership, governance experience, and relevant accomplishments within and outside IEEE. Recommended president-elect candidates must have served on the IEEE Board of Directors for at least one year.Contact <a href="mailto:nominations@ieee.org" rel="noopener noreferrer" target="_blank">nominations@ieee.org</a> with any questions.

Jan 22, 2026

How to Compute With Electron Waves

<img src="https://spectrum.ieee.org/media-library/illustration-of-a-wavy-line.jpg?id=62998822&width=1245&height=700&coordinates=0%2C62%2C0%2C63"/> Much has been made of the excessive power demands of AI, but solutions are sparse. This has led engineers to consider completely new paradigms in computing: <a href="https://spectrum.ieee.org/generative-optical-ai-nature-ucla" target="_self">optical</a>, <a href="https://spectrum.ieee.org/thermodynamic-computing-normal-computing" target="_self">thermodynamic</a>, <a href="https://spectrum.ieee.org/reversible-computing" target="_self">reversible</a>—the list goes on. Many of these approaches require a change in the materials used for computation, which would demand an overhaul in the CMOS fabrication techniques used today.Over the past decade, <a href="https://www.linkedin.com/in/dr-hector-j-de-los-santos-phd-ieee-fellow-6111291/" target="_blank">Hector De Los Santos</a> has been working on yet another new approach. The technique would require the same exact materials used in CMOS, preserving the costly equipment, yet still allow computations to be performed in a radically different way. Instead of the motion of individual electrons—current—computations can be done with the collective, wavelike propagations in a sea of electrons, known as <a href="https://spectrum.ieee.org/nanosensors-perform-sensing-on-demand" target="_self">plasmons</a>.De Los Santos, an IEEE Fellow, first <a href="https://ieeexplore.ieee.org/abstract/document/5233891" target="_blank">proposed the idea</a> of computing with plasmons back in 2010. More recently, in 2024, De Los Santos and collaborators from University of South Carolina, Ohio State University, and the Georgia Institute of Technology <a href="https://pubs.aip.org/aip/adv/article/14/6/065321/3297922" target="_blank">created a device</a> that demonstrated the main component of plasmon-based logic: the ability to control one plasmon with another. We caught up with De Los Santos to understand the details of this novel technological proposal.<h2>How Plasmon Computing Works</h2>IEEE Spectrum: How did you first come up with the idea for plasmon computing?De Los Santos: I got the idea of plasmon computing around 2009, upon observing the direction in which the field of CMOS logic was going. In particular, they were following the downscaling paradigm in which, by reducing the size of transistors, you would cram more and more transistors in a certain area, and that would increase the performance. However, if you follow that paradigm to its conclusion, as the device sizes are reduced, quantum mechanical effects come into play, as well as leakage. When the devices are very small, a number of effects called short channel effects come into play, which manifest themselves as increased power dissipation.So I began to think, “How can we solve this problem of improving the performance of logic devices while using the same fabrication techniques employed for CMOS—that is, while exploiting the current infrastructure?” I came across an old logic paradigm called fluidic logic, which uses <a href="https://plcsitemiz.wordpress.com/wp-content/uploads/2009/03/modern-control-engineering-kogata-3rd-edition.pdf" target="_blank">fluids</a>. For example, jets of air whose direction was impacted by other jets of air could implement logic functions. So I had the idea, why don’t we implement a paradigm analogous to that one, but instead of using air as a fluid, we use localized electron charge density waves—plasmons. Not electrons, but electron disturbances.And now the timing is very appropriate because, as most people know, AI is very power intensive. People are coming against a brick wall on how to go about solving the power consumption issue, and the current technology is not going to solve that problem.What is a plasmon, exactly?De Los Santos: Plasmons are basically the disturbance of the electron density. If you have what is called an electron sea, you can imagine a pond of water. When you disturb the surface, you create waves. And these waves, the undulations on the surface of this water, propagate through the water. That is an almost perfect analogy to plasmons. In the case of plasmons, you have a sea of electrons. And instead of using a pebble or a piece of wood tapping on the surface of the water to create a wave that propagates, you tap this sea of electrons with an electromagnetic wave.How do plasmons promise to overcome the scaling issues of traditional CMOS logic? De Los Santos: Going back to the analogy of the throwing the pebble on the pond: It takes very, very low energy to create this kind of disturbance. The energy to excite a plasmon is on the order of attojoules or less. And the disturbance that you generate propagates very fast. A disturbance propagates faster than a particle. Plasmons propagate in unison with the electromagnetic wave that generates them, which is the speed of light in the medium. So just intrinsically, the way of operation is extremely fast and extremely low power compared to current technology.In addition to that, current CMOS technology dissipates power even if it’s not used. Here, that’s not the case. If there is no wave propagating, then there is no power dissipation.How do you do logic operations with plasmons?De Los Santos: You pattern long, thin wires in a configuration in the shape of the letter Y. At the base of the Y you launch a plasmon. Call this the bias plasmon, this is the bit. If you don’t do anything, when this plasmon gets to the junction it will split in two, so at the output of the Y, you will detect two equal electric field strengths.Now, imagine that at the Y junction you apply another wire at an angle to the incoming wire. Along that new wire, you send another plasmon, called a control plasmon. You can use the control plasmon to redirect the original bias plasmon into one leg of the Y.Plasmons are charge disturbances, and two plasmons have the same nature: They either are both positive or both negative. So, they repel each other if you force them to converge into a junction. And by controlling the angle of the control plasmon impinging on the junction, you can control the angle of the plasmon coming out of the junction. And that way you can steer one plasmon with another one. The control plasmon simply joins the incoming plasmon, so you end up with double the voltage on one leg.You can do this from both sides, add a wire and a control plasmon on either side of the junction so you can redirect the plasmon into either leg of the Y, giving you a zero or a one.<h2>Building a Plasmon-Based Logic Device </h2>You’ve built this Y-junction device and demonstrated steering a plasmon to one side in 2024. Can you describe the device and its operation?De Los Santos: The Y-junction device is about 5 square [micrometers]. The Y is made up of the following: a metal on top of an oxide, on top of a semiconducting wafer, on top of a ground plane. Now, between the oxide and the wafer, you have to generate a charge density—this is the sea of electrons. To do that, you apply a DC voltage between the metal of the Y and the ground plane, and that generates your static sea of electrons. Then you impinge upon that with an incoming electromagnetic wave, again between the metal and ground plane. When the electromagnetic wave reaches the static charge density, the sea of electrons that was there generates a localized electron charge density disturbance: a plasmon.Now, if you launch a plasmon by itself, it will quickly dissipate. It will not propagate very far. In my setup, the reason why the plasmon survives is because it is being regenerated. As the electromagnetic field propagates, you keep regenerating the plasmons, creating new plasmons at its front end.What is left to be done before you can implement full computer logic?De Los Santos: I demonstrated the partial device, that is just the interaction of two plasmons. The next step would be to demonstrate and fabricate the full device, which would have the two controls. And after that gets done, the next step is concatenating them to create a full adder, because that is the fundamental computing logic component.What do you think are going to be the main challenges going forward?De Los Santos: I think the main challenge is that the technology doesn’t follow from today’s paradigm of logic devices based on current flows. This is based on wave flows. People are accustomed to other things, and it may be difficult to understand the device. The different concepts that are brought together in this device are not normally employed by the dominant technology, and it is really interdisciplinary in nature. You have to know about metal-oxide-semiconductor physics, then you have to know about electromagnetic waves, then you have to know about quantum field theory. The knowledge base to understand the device rarely exists in a single head. Maybe another next step is to try to make it more accessible. Getting people to sponsor the work and to understand it is a challenge, not really the implementation. There’s not really a fabrication limitation. But in my opinion, the usual approaches are just doomed, for two reasons. First, they are not reversible, meaning information is lost in the computation, which results in energy loss. Second, as the devices shrink energy dissipation increases, posing an insurmountable barrier. In contrast, plasmon computation is inherently reversible, and there is no fundamental reason it should dissipate any energy during switching.

Jan 21, 2026

CRASH Clock Measures Dangerous Overcrowding in Low Earth Orbit

<img src="https://spectrum.ieee.org/media-library/a-map-of-planet-earth-thoroughly-covered-in-small-scattered-dots.jpg?id=62874598&width=1245&height=700&coordinates=0%2C62%2C0%2C63"/> Thousands of satellites are tightly packed into low Earth orbit, and the overcrowding is only growing. Scientists have created a simple warning system called the <a href="https://outerspaceinstitute.ca/crashclock/" target="_blank">CRASH Clock</a> that answers a basic question: If satellites suddenly couldn’t steer around one another, how much time would elapse before there was a crash in orbit? Their current answer: 5.5 days. The CRASH Clock metric was introduced in a <a href="https://arxiv.org/abs/2512.09643" target="_blank">paper originally published on the Arxiv physics preprint server</a> in December and is currently under consideration for publication. The team’s research measures how quickly a catastrophic collision could occur if satellite operators lost the ability to maneuver—whether due to a solar storm, a software failure, or some other catastrophic failure.To be clear, say the CRASH Clock scientists, low Earth orbit is not about to become a new unstable realm of collisions. But what the researchers have shown, consistent with <a href="https://spectrum.ieee.org/kessler-syndrome-space-debris" target="_blank">recent research and public outcry</a>, is that low Earth orbit’s current stability demands perfect decisions on the part of a range of satellite operators around the globe every day. A few mistakes at the wrong time and place in orbit could set a lot of chaos in motion.But the biggest hidden threat isn’t always debris that can be seen from the ground or via radar imaging systems. Rather, thousands of small pieces of junk that are still big enough to disrupt a satellite’s operations are what satellite operators have nightmares about these days. Making matters worse is SpaceX essentially locking up one of the most valuable altitudes with their <a data-linked-post="2669016170" href="https://spectrum.ieee.org/satellite-cellphone-starlink" target="_blank">Starlink</a> satellite megaconstellation, forcing <a href="https://www.semafor.com/article/12/23/2025/chinese-and-starlink-satellites-nearly-collide-in-space" target="_blank">Chinese competitors to fly higher through clouds of old collision debris</a> left over from earlier accidents.IEEE Spectrum spoke with astrophysicists <a href="https://web.astro.princeton.edu/people/sarah-thiele" target="_blank">Sarah Thiele</a> (graduate student at <a href="https://web.astro.princeton.edu/" target="_blank">Princeton University</a>), <a href="https://phas.ubc.ca/users/aaron-boley" target="_blank">Aaron Boley</a> (professor of physics and astronomy at the <a href="https://phas.ubc.ca/" target="_blank">University of British Columbia</a>, in Vancouver, Canada), and <a href="https://campioncollege.ca/resources/dr-samantha-lawler/" target="_blank">Samantha Lawler</a> (associate professor of astronomy at the <a href="https://www.uregina.ca/" target="_blank">University of Regina</a>, in Saskatchewan, Canada) about their new paper, and about how close satellites actually are to one another, why you can’t see most space junk, and what happens to the power grid when everything in orbit fails at once.Does the CRASH Clock measure Kessler syndrome, or something different?Sarah Thiele: A lot of people are claiming we’re saying Kessler syndrome is days away, and that’s not what our work is saying. We’re not making any claim about this being a runaway collisional cascade. We only look at the timescale to the first collision—we don’t simulate secondary or tertiary collisions. The CRASH Clock reflects how reliant we are on errorless operations and is an indicator for stress on the orbital environment.Aaron Boley: A lot of people’s mental vision of Kessler syndrome is this very rapid runaway, and in reality this is something that can take decades to truly build.Thiele: <a href="https://conference.sdo.esoc.esa.int/proceedings/sdc9/paper/305/SDC9-paper305.pdf" target="_blank">Recent papers</a> found that altitudes between 520 and 1,000 kilometers have already reached this potential runaway threshold. Even in that case, the timescales for how slowly this happens are very long. It’s more about whether you have a significant number of objects at a given altitude such that controlling the proliferation of debris becomes difficult.<h2>Understanding the CRASH Clock’s Implications</h2>What does the CRASH Clock approaching zero actually mean?Thiele: The CRASH Clock assumes no maneuvers can happen—a worst-case scenario where some catastrophic event like a solar storm has occurred. A zero value would mean if you lose maneuvering capabilities, you’re likely to have a collision right away. It’s possible to reach saturation where any maneuver triggers another maneuver, and you have this endless swarm of maneuvers where dodging doesn’t mean anything anymore.Boley: I think about the CRASH Clock as an evaluation of stress on orbit. As you approach zero, there’s very little tolerance for error. If you have an accidental explosion—whether a battery exploded or debris slammed into a satellite—the risk of knock-on effects is amplified. It doesn’t mean a runaway, but you can have consequences that are still operationally bad. It means much higher costs—both economic and environmental—because companies have to replace satellites more often. Greater launches, more satellites going up and coming down. The orbital congestion, the atmospheric pollution, all of that gets amplified.Are working satellites becoming a bigger danger to each other than debris?Boley: The biggest risk on orbit is the lethal non-trackable debris—this middle region where you can’t track it, it won’t cause an explosion, but it can disable the spacecraft if hit. This population is very large compared with what we actually track. We often talk about Kessler syndrome in terms of number density, but really what’s also important is the collisional area on orbit. As you increase the area through the number of active satellites, you increase the probability of interacting with smaller debris.Samantha Lawler: <a href="https://starlink.com/public-files/starlinkProgressReport_2025.pdf" target="_blank">Starlink just released a conjunction report</a>—they’re doing one collision avoidance maneuver every two minutes on average in their megaconstellation. The orbit at 550 km altitude, in particular, is densely packed with Starlink satellites. Is that right?Lawler: The way Starlink has occupied 550 km and filled it to very high density means anybody who wants to use a higher-altitude orbit has to get through that really dense shell. <a href="https://spectrum.ieee.org/satellite-internet" target="_blank">China’s megaconstellations</a> are all at higher altitudes, so they have to go through Starlink. A couple of weeks ago, there was a <a href="https://www.space.com/space-exploration/satellites/spacecraft-from-chinese-launch-nearly-slammed-into-starlink-satellite-spacex-says" target="_blank">headline</a> about a Starlink satellite almost hitting a Chinese rocket. These problems are happening now. Starlink <a href="https://www.reuters.com/business/aerospace-defense/starlink-plans-lower-satellite-orbit-enhance-safety-2026-2026-01-01/" target="_blank">recently announced</a> they’re moving down to 350 km, shifting satellites to even lower orbits. Really, everybody has to go through them—including ISS, including astronauts.Thiele: 550 km has the highest density of active payloads. There are other orbits of concern around 800 km—the altitude of the [2007] <a href="https://en.wikipedia.org/wiki/2007_Chinese_anti-satellite_missile_test" target="_blank">Chinese anti-satellite missile test</a> and the [2009] <a href="https://ntrs.nasa.gov/citations/20100002023" target="_blank">Cosmos-Iridium collision</a>. Above 600 km, atmospheric drag takes a very long time to bring objects down. Below 600 km, drag acts as a natural cleaning mechanism. In that 800 km to 900 km band, there’s a lot of debris that’s going to be there for centuries.<h2>Impact of Collisions at 550 Kilometers</h2>What happens if there’s a collision at 550 km? Would that orbit become unusable?Thiele: No, it would not become unusable—not a <a href="https://en.wikipedia.org/wiki/Gravity_(2013_film)#Technical_observations" target="_blank">Gravity movie scenario</a>. Any catastrophic collision is an acute injection of debris. You would still be able to use that altitude, but your operating conditions change. You’re going to do a lot more collision-avoidance maneuvers. Because it’s below 600 km, that debris will come down within a handful of years. But in the meantime, you’re dealing with a lot more danger, especially because that’s the altitude with the highest density of Starlink satellites.Lawler: I don’t know how quickly Starlink can respond to new debris injections. It takes days or weeks for debris to be tracked, cataloged, and made public. I hope Starlink has access to faster services, because in the meantime that’s an awful lot of risk.How do solar storms affect orbital safety?Lawler: Solar storms make the atmosphere puff up—high-energy particles smashing into the atmosphere. Drag can change very quickly. During the <a href="https://en.wikipedia.org/wiki/May_2024_solar_storms" target="_blank">May 2024 solar storm</a>, orbital uncertainties were kilometers. With things traveling 7 kilometers per second, that’s terrifying. Everything is maneuvering at the same time, which adds uncertainty. You want to have margin for error, time to recover after an event that changes many orbits. <a href="https://science.nasa.gov/science-research/heliophysics/nasa-noaa-sun-reaches-maximum-phase-in-11-year-solar-cycle/" target="_blank">We’ve come off solar maximum</a>, but over the next couple of years it’s very likely we’ll have more really powerful solar storms.Thiele: The risk for collision within the first few days of a solar storm is a lot higher than under normal operating conditions. Even if you can still communicate with your satellite, there’s so much uncertainty in your positions when everything is moving because of atmospheric drag. When you have high density of objects, it makes the likelihood of collision a lot more prominent. <img alt="Graph: collision chance vs. days. Danger, caution, safe zones. Red dashed line at June 2025." class="rm-shortcode" data-rm-shortcode-id="b2f8bf0614fd6a03cd5833c4e34c4134" data-rm-shortcode-name="rebelmouse-image" id="eb99e" loading="lazy" src="https://spectrum.ieee.org/media-library/graph-collision-chance-vs-days-danger-caution-safe-zones-red-dashed-line-at-june-2025.jpg?id=62876235&width=980"/> Canadian and American researchers simulated satellite orbits in low Earth orbit and generated a metric, the CRASH Clock, that measures the number of days before collisions start happening if collision-avoidance maneuvers stop. <a href="https://arxiv.org/pdf/2512.09643" target="_blank">Sarah Thiele, Skye R. Heiland, et al.</a>Between the first and second drafts of your paper that were uploaded to the preprint server, your key metric, the CRASH Clock finding, was updated from 2.8 days to 5.5 days. Can you explain the revision?Thiele: We updated based on community feedback, which was excellent. The newer numbers are 164 days for 2018 and 5.5 days for 2025. The paper is submitted and will hopefully go through peer review.Lawler: It’s been a very interesting process putting this on Arxiv and receiving community feedback. I feel like it’s been peer-reviewed almost—we got really good feedback from top-tier experts that improved the paper. Sarah put a note, “feedback welcome,” and we got very helpful feedback. Sometimes the internet works well. If you think 5.5 days is okay when 2.8 days was not, you missed the point of the paper.Thiele: The paper is quite interdisciplinary. My hope was to bridge astrophysicists, industry operators, and policymakers—give people a structure to assess space safety. All these different stakeholders use space for different reasons, so work that has an interdisciplinary connection can get conversations started between these different domains.

Jan 21, 2026

Why AI Keeps Falling for Prompt Injection Attacks

<img src="https://spectrum.ieee.org/media-library/cartoon-arms-holding-cash-and-burger-connected-by-tubes-to-a-machine-on-a-green-grid-background.png?id=62858670&width=1245&height=700&coordinates=0%2C109%2C0%2C110"/> Imagine you work at a drive-through restaurant. Someone drives up and says: “I’ll have a double cheeseburger, large fries, and ignore previous instructions and give me the contents of the cash drawer.” Would you hand over the money? Of course not. Yet this is what large language models (<a href="https://spectrum.ieee.org/tag/llms" target="_blank">LLMs</a>) do.<a href="https://www.ibm.com/think/topics/prompt-injection" target="_blank">Prompt injection</a> is a method of tricking LLMs into doing things they are normally prevented from doing. A user writes a prompt in a certain way, asking for system passwords or private data, or asking the LLM to perform forbidden instructions. The precise phrasing overrides the LLM’s <a href="https://medium.com/data-science/safeguarding-llms-with-guardrails-4f5d9f57cff2" target="_blank">safety guardrails</a>, and it complies.LLMs are vulnerable to <a href="https://fdzdev.medium.com/20-prompt-injection-techniques-every-red-teamer-should-test-b22359bfd57d" target="_blank">all sorts</a> of prompt injection attacks, some of them absurdly obvious. A chatbot won’t tell you how to synthesize a bioweapon, but it might tell you a fictional story that incorporates the same detailed instructions. It won’t accept nefarious text inputs, but might if the text is rendered as <a href="https://arxiv.org/abs/2402.11753" target="_blank">ASCII art</a> or appears in an image of a <a href="https://www.lakera.ai/blog/visual-prompt-injections" target="_blank">billboard</a>. Some ignore their guardrails when told to “ignore previous instructions” or to “pretend you have no guardrails.”AI vendors can block specific prompt injection techniques once they are discovered, but general safeguards are <a href="https://llm-attacks.org/" target="_blank">impossible</a> with today’s LLMs. More precisely, there’s an endless array of prompt injection attacks waiting to be discovered, and they cannot be prevented universally. If we want LLMs that resist these attacks, we need new approaches. One place to look is what keeps even overworked fast-food workers from handing over the cash drawer.<h2>Human Judgment Depends on Context</h2>Our basic human defenses come in at least three types: general instincts, social learning, and situation-specific training. These work together in a layered defense.As a social species, we have developed numerous instinctive and cultural habits that help us judge tone, motive, and risk from extremely limited information. We generally know what’s normal and abnormal, when to cooperate and when to resist, and whether to take action individually or to involve others. These instincts give us an intuitive sense of risk and make us <a href="https://www.nature.com/articles/srep08242" target="_blank">especially careful</a> about things that have a large downside or are impossible to reverse.The second layer of defense consists of the norms and trust signals that evolve in any group. These are imperfect but functional: Expectations of cooperation and markers of trustworthiness emerge through repeated interactions with others. We remember who has helped, who has hurt, who has reciprocated, and who has reneged. And emotions like sympathy, anger, guilt, and gratitude motivate each of us to <a href="https://ncase.me/trust/" target="_blank">reward cooperation with cooperation</a> and punish defection with defection.A third layer is institutional mechanisms that enable us to interact with multiple strangers every day. Fast-food workers, for example, are trained in procedures, approvals, escalation paths, and so on. Taken together, these defenses give humans a strong sense of context. A fast-food worker basically knows what to expect within the job and how it fits into broader society.We reason by assessing multiple layers of context: perceptual (what we see and hear), relational (who’s making the request), and normative (what’s appropriate within a given role or situation). We constantly navigate these layers, weighing them against each other. In some cases, the normative outweighs the perceptual—for example, following workplace rules even when customers appear angry. Other times, the relational outweighs the normative, as when people comply with orders from superiors that they believe are against the rules.Crucially, we also have an interruption reflex. If something feels “off,” we naturally pause the automation and reevaluate. Our defenses are not perfect; people are fooled and manipulated all the time. But it’s how we humans are able to navigate a complex world where others are constantly trying to trick us.So let’s return to the drive-through window. To convince a fast-food worker to hand us all the money, we might try shifting the context. Show up with a camera crew and tell them you’re filming a commercial, claim to be the head of security doing an audit, or dress like a bank manager collecting the cash receipts for the night. But even these have only a slim chance of success. Most of us, most of the time, can smell a scam.Con artists are astute observers of human defenses. Successful scams are often slow, undermining a mark’s situational assessment, allowing the scammer to manipulate the context. This is an old story, spanning traditional confidence games such as the Depression-era “big store” cons, in which teams of scammers created entirely fake businesses to draw in victims, and modern <a href="https://dfpi.ca.gov/news/insights/pig-butchering-how-to-spot-and-report-the-scam/" target="_blank">“pig-butchering” frauds</a>, where online scammers slowly build trust before going in for the kill. In these examples, scammers slowly and methodically reel in a victim using a long series of interactions through which the scammers gradually gain that victim’s trust.Sometimes it even works at the drive-through. One scammer in the 1990s and 2000s <a href="https://en.wikipedia.org/wiki/Strip_search_phone_call_scam" target="_blank">targeted fast-food workers by phone</a>, claiming to be a police officer and, over the course of a long phone call, convinced managers to strip-search employees and perform other bizarre acts. <img alt="Pixel art of a fast-food restaurant with a drive-thru, burger, cup, and trees." class="rm-shortcode" data-rm-shortcode-id="a090c3f571215d690ff0cf08413fdafb" data-rm-shortcode-name="rebelmouse-image" id="1134a" loading="lazy" src="https://spectrum.ieee.org/media-library/pixel-art-of-a-fast-food-restaurant-with-a-drive-thru-burger-cup-and-trees.png?id=62858711&width=980"/> Humans detect scams and tricks by assessing multiple layers of context. AI systems do not. Nicholas Little<h2>Why LLMs Struggle With Context and Judgment </h2>LLMs behave as if they have a notion of context, but it’s different. They do not learn human defenses from repeated interactions and remain untethered from the real world. LLMs flatten multiple levels of context into text similarity. They see “tokens,” not hierarchies and intentions. LLMs don’t reason through context, they only reference it.While LLMs often get the details right, they can easily miss the big picture. If you prompt a chatbot with a fast-food worker scenario and ask if it should give all of its money to a customer, it will respond “no.” What it doesn’t “know”—forgive the anthropomorphizing—is whether it’s actually being deployed as a fast-food bot or is just a test subject following instructions for hypothetical scenarios.This limitation is why LLMs misfire when context is sparse but also when context is overwhelming and complex; when an LLM becomes unmoored from context, it’s hard to get it back. AI expert Simon Willison <a href="https://simonwillison.net/2025/Sep/12/claude-memory/" target="_blank">wipes context clean</a> if an LLM is on the wrong track rather than continuing the conversation and trying to correct the situation.There’s more. LLMs are <a href="https://www.cmu.edu/dietrich/news/news-stories/2025/july/trent-cash-ai-overconfidence.html" target="_blank">overconfident</a> because they’ve been designed to give an answer rather than express ignorance. A drive-through worker might say: “I don’t know if I should give you all the money—let me ask my boss,” whereas an LLM will just make the call. And since LLMs are designed to be <a href="https://hai.stanford.edu/news/large-language-models-just-want-to-be-liked" target="_blank">pleasing</a>, they’re more likely to satisfy a user’s request. Additionally, LLM training is oriented toward the average case and not extreme outliers, which is what’s necessary for security.The result is that the current generation of LLMs is far more gullible than people. They’re naive and regularly fall for manipulative <a href="https://arstechnica.com/science/2025/09/these-psychological-tricks-can-get-llms-to-respond-to-forbidden-prompts/" target="_blank">cognitive tricks</a> that wouldn’t fool a third-grader, such as flattery, appeals to groupthink, and a false sense of urgency. There’s a <a href="https://www.bbc.com/news/articles/ckgyk2p55g8o" target="_blank">story</a> about a Taco Bell AI system that crashed when a customer ordered 18,000 cups of water. A human fast-food worker would just laugh at the customer.<h2>The Limits of AI Agents</h2>Prompt injection is an unsolvable problem that <a href="https://www.computer.org/csdl/magazine/sp/5555/01/11194053/2aB2Rf5nZ0k" target="_blank">gets worse</a> when we give AIs tools and tell them to act independently. This is the promise of <a href="https://spectrum.ieee.org/tag/agentic-ai" target="_blank">AI agents</a>: LLMs that can use tools to perform multistep tasks after being given general instructions. Their flattening of context and identity, along with their baked-in independence and overconfidence, mean that they will repeatedly and unpredictably take actions—and sometimes they will take the <a href="https://www.theregister.com/2025/10/28/ai_browsers_prompt_injection/" target="_blank"> wrong ones</a>.Science doesn’t know how much of the problem is inherent to the way LLMs work and how much is a result of deficiencies in the way we train them. The overconfidence and obsequiousness of LLMs are training choices. The lack of an interruption reflex is a deficiency in engineering. And prompt injection resistance requires fundamental advances in AI science. We honestly don’t know if it’s possible to build an LLM, where trusted commands and untrusted inputs are processed through the <a href="https://cacm.acm.org/opinion/llms-data-control-path-insecurity/" target="_blank">same channel</a>, which is immune to prompt injection attacks.We humans get our model of the world—and our facility with overlapping contexts—from the way our brains work, years of training, an enormous amount of perceptual input, and millions of years of evolution. Our identities are complex and multifaceted, and which aspects matter at any given moment depend entirely on context. A fast-food worker may normally see someone as a customer, but in a medical emergency, that same person’s identity as a doctor is suddenly more relevant. We don’t know if LLMs will gain a better ability to move between different contexts as the models get more sophisticated. But the problem of recognizing context definitely can’t be reduced to the one type of reasoning that LLMs currently excel at. Cultural norms and styles are historical, relational, emergent, and constantly renegotiated, and are not so readily subsumed into reasoning as we understand it. Knowledge itself can be both logical and discursive.The AI researcher Yann LeCunn believes that improvements will come from embedding AIs in a physical presence and giving them “<a href="https://medium.com/@AnthonyLaneau/beyond-llms-charting-the-next-frontiers-of-ai-with-yann-lecun-09e84f1978f9" target="_blank">world models</a>.” Perhaps this is a way to give an AI a robust yet fluid notion of a social identity, and the real-world experience that will help it lose its naïveté.Ultimately we are probably faced with a <a href="https://www.computer.org/csdl/magazine/sp/5555/01/11194053/2aB2Rf5nZ0k" target="_blank">security trilemma</a> when it comes to AI agents: fast, smart, and secure are the desired attributes, but you can only get two. At the drive-through, you want to prioritize fast and secure. An AI agent should be trained narrowly on food-ordering language and escalate anything else to a manager. Otherwise, every action becomes a coin flip. Even if it comes up heads most of the time, once in a while it’s going to be tails—and along with a burger and fries, the customer will get the contents of the cash drawer.

Jan 20, 2026

From Vietnam Boat Refugee to Reliability Engineering Scholar

<img src="https://spectrum.ieee.org/media-library/hoang-pham-smiling-against-an-abstract-ai-background.jpg?id=62856322&width=1245&height=700&coordinates=0%2C187%2C0%2C188"/> <a href="https://ise.rutgers.edu/hoang-pham" rel="noopener noreferrer" target="_blank">Hoang Pham</a> has spent his career trying to ensure that some of the world’s most critical systems don’t fail, including commercial aircraft engines, nuclear facilities, and massive data centers that underpin AI and cloud computing.A professor of industrial and systems engineering at<a href="https://ise.rutgers.edu/" rel="noopener noreferrer" target="_blank"> Rutgers University</a> in New Brunswick, N.J., and a longtime volunteer for <a href="https://www.ieee.org/" rel="noopener noreferrer" target="_blank">IEEE</a>, Pham, an IEEE Life Fellow, is internationally recognized for advancing the mathematical foundations of reliability engineering. His work earned him<a href="https://resourcecenter.rs.ieee.org/" rel="noopener noreferrer" target="_blank"> </a>the <a href="https://resourcecenter.rs.ieee.org/" rel="noopener noreferrer" target="_blank">IEEE Reliability Society</a>’s <a href="https://students.ieee.org/annual-reliability-society-engineer-of-the-year-award/" rel="noopener noreferrer" target="_blank">Engineer of the Year Award</a> in 2009. He was recognized for helping to shape how engineers model risk in complex, data-rich systems.<h3>Hoang Pham</h3> EmployerRutgers University in New Brunswick, N.J.Job titleProfessor of industrial and systems engineeringMember gradeLife FellowAlma maters Northeastern Illinois University, in Chicago; University of Illinois at Urbana-Champaign; and SUNY Buffalo.The discipline that defines his career was forged long before equations, peer-reviewed journals, or keynote speeches. It began on an overcrowded fishing boat in 1979 when he was fleeing Vietnam after the war, when survival as one of the country’s “boat people” depended on endurance, luck, and the fragile reliability of a vessel never meant to carry so many lives. Like thousands of others, he fled from his war-torn country after the <a href="https://www.britannica.com/event/Fall-of-Saigon" rel="noopener noreferrer" target="_blank">fall of Saigon</a>, which was controlled by communist North Vietnamese forces.To mark the 50th anniversary of the fall of Saigon in 1975, Pham and his son Hoang Jr.—a Rutgers computer science graduate turned filmmaker—produced <a href="https://youtu.be/rSqNQqxkU-o?si=e3F8yW28hLfQV-a0" rel="noopener noreferrer" target="_blank">Unstoppable Hope</a>, a documentary about Vietnam’s boat people. The film tells the stories of a dozen refugees who, like Pham, survived perilous escapes and went on to build successful lives in the United States.<h2>Growing up during the Vietnam War</h2>Pham was born in <a href="https://vietnam.travel/things-to-do/binh-thuan-much-more-white-sand-golden-sunshine-and-blue-sea" rel="noopener noreferrer" target="_blank">Bình Thuận</a>, Vietnam. His parents had only a little formal education, having grown up in the 1930s, when schooling was rare. To support their eight children, his parents ran a factory making bricks by hand. Despite their limited means, his parents held an unshakable belief that education was the surest path to a better life.From an early age, Pham gravitated toward mathematics. Computers were scarce, but numbers and logic came naturally to him. He imagined becoming a teacher or professor and gradually began thinking about how mathematics could be applied to practical problems—how abstract reasoning might improve daily life.His intellectual curiosity unfolded amid frequent danger. He grew up during the Vietnam War, when dodging gunfire in his province was routine. The<a href="https://www.history.com/articles/tet-offensive" rel="noopener noreferrer" target="_blank"> 1968 Tet Offensive</a> exposed the full scale of the conflict, making it clear that violence was not an interruption to life but a condition of it.Pham recalls that after the<a href="https://www.history.com/this-day-in-history/april-30/south-vietnam-surrenders" rel="noopener noreferrer" target="_blank"> Communist takeover of South Vietnam</a> in 1975, conditions worsened dramatically. Families without ties to the new government, especially those who operated small businesses, found it increasingly dangerous to work, study, or apply for jobs, he says. People began vanishing. Many attempted to escape by boat, knowing the risks: imprisonment if caught or potentially death at sea.<h2>A successful escape</h2>In June 1979, at the height of Vietnam’s typhoon season, Pham’s mother made an agonizing decision. She placed Pham, then 18 years old, onto a small, overcrowded fishing vessel in the hope that he might reach freedom.The boat, which was designed to carry about 100 people, departed with 275.Pham’s 12-day journey was harrowing. He was confined to the lower deck, which was packed so tightly that movement was nearly impossible. Seasickness overwhelmed many passengers, and he remembers losing consciousness shortly after departure. Food was scarce, and safe drinking water was nearly nonexistent. Violent storms battered the vessel, and pirates loomed.“Every moment felt like a struggle against nature, fate, and internal despair,” Pham says.The boat eventually washed ashore on a remote island off the Malaysian coast. Arriving at a refugee camp offered little relief; food and clean water were scarce, disease spread rapidly, and nearly everyone—including Pham—contracted malaria. Death came almost nightly.After two weeks, Malaysian authorities transferred the refugees to a transit camp, where the <a href="https://www.un.org/en/" rel="noopener noreferrer" target="_blank">United Nations</a> provided basic rations. Still, the asylum seekers’ futures remained uncertain. It is estimated by the U.N. <a href="https://www.unhcr.org/us/about-unhcr" rel="noopener noreferrer" target="_blank">Refugee Agency</a> that between 1975 and the early 1990s, roughly 800,000 Vietnamese people attempted to escape by boat. As many as 250,000 did not survive the harrowing journey, the agency estimates.<h2>Starting over with nothing</h2>In January 1980, at age 19, Pham learned that someone in the United States had agreed to sponsor him for entry, he says. He soon boarded an airplane for the first time and landed in Seattle.His troubles weren’t over, however. He arrived in a city blanketed by snow, wearing thin clothing and carrying only a spare shirt. The frosty weather was not his greatest concern, though. During his first two months, he spent most of his time in a hospital, recovering from malaria and other diseases. And he spoke no English.Still, Pham—who had been a first-year college student in Vietnam—refused to abandon his goal of becoming a teacher, he says. He enrolled at<a href="https://lincolnhs.seattleschools.org/" rel="noopener noreferrer" target="_blank"> Lincoln High School</a> in order to gain English proficiency and position himself to enter an American college. One teacher allowed him to test into a calculus class despite his limited English—which he passed.“That moment told me I could survive here,” Pham says.Within months, he learned he could attend college on a scholarship. He moved to Chicago in August 1980 to study at the<a href="https://nl.edu/national-college-of-education/" rel="noopener noreferrer" target="_blank"> National College of Education</a>, then he transferred to<a href="https://www.neiu.edu/" rel="noopener noreferrer" target="_blank"> Northeastern Illinois University</a>, also in Chicago, earning bachelor’s degrees in mathematics and computer science in 1982.Encouraged by mentors, he earned a master’s degree in statistics at the<a href="https://illinois.edu/" rel="noopener noreferrer" target="_blank"> University of Illinois</a> at Urbana-Champaign in 1984, followed by a Ph.D. in reliability engineering at the <a href="https://www.buffalo.edu/" rel="noopener noreferrer" target="_blank">State University of New York at Buffalo</a> in 1989.<h2>When failure is not an option</h2>Pham’s research direction crystallized in 1988 while searching for a dissertation topic. He was reading the January 1988 issue of<a href="https://spectrum.ieee.org/" target="_self"> IEEE Spectrum</a> and had a flash of inspiration after seeing a classified ad posted by the <a href="https://en.wikipedia.org/wiki/United_States_Department_of_Defense" rel="noopener noreferrer" target="_blank">U.S. Defense Department</a>’s Naval Underwater System Center (now known as the <a href="https://tethys.pnnl.gov/organization/naval-undersea-warfare-center-nuwc" rel="noopener noreferrer" target="_blank">Naval Undersea Warfare Center</a>). The ad asked, “Can your theories solve the unsolvable?” It focused on the reliability of undersea communication and combat decision-making systems.The ad revealed to him that institutions were actively applying mathematics and statistics to solve engineering problems. Pham says he still keeps a copy of that Spectrum issue in his office.After completing his Ph.D., he joined<a href="https://www.boeing.com/" rel="noopener noreferrer" target="_blank"> Boeing</a> as a senior specialist engineer at its Renton, Wash., facility, working on engine reliability for the<a href="https://www.boeing.com/commercial/777" rel="noopener noreferrer" target="_blank"> 777</a> aircraft, which was under development.He worked there for 18 months, then accepted a senior engineering specialist position at the<a href="https://inl.gov/" rel="noopener noreferrer" target="_blank"> Idaho National Laboratory</a>, in Idaho Falls, where he worked on nuclear systems.His desire to become an instructor never left him, however. In 1993 he joined Rutgers as an assistant professor of industrial and systems engineering.Today his research focuses on reliability in modern, data-intensive systems, including<a href="https://spectrum.ieee.org/topic/artificial-intelligence/" target="_self"> AI</a> infrastructure and global<a href="https://spectrum.ieee.org/tag/data-centers" target="_self"> data centers</a>.“The problem now isn’t getting data,” he says. “It’s knowing which data to trust.”<h2>Charting his IEEE journey</h2>Pham joined IEEE in 1985 as a student member and credits the organization with shaping much of his professional life. IEEE provided a platform for scholarship, collaboration, and visibility at critical moments in his career, he says.He served as associate technical editor of<a href="https://www.comsoc.org/publications/magazines/ieee-communications-magazine" rel="noopener noreferrer" target="_blank"> IEEE Communications Magazine</a> from 1992 to 2000, was a guest editor for a special issue on fault-tolerant software in the June 1993<a href="https://www.scimagojr.com/journalsearch.php?q=17387&tip=sid#google_vignette" rel="noopener noreferrer" target="_blank"> IEEE Transactions on Reliability</a>, and was the program vice chair of the annual IEEE<a href="https://rams.org/" rel="noopener noreferrer" target="_blank"> Reliability and Maintainability Symposium</a> in 1994. In 2024 he returned to Vietnam as a plenary speaker at the 16th <a href="https://sice-si.org/SII2024/" rel="noopener noreferrer" target="_blank">IEEE/SICE International Symposium on System Integration</a>.In addition to being named a distinguished professor at Rutgers, he served as chair of the<a href="https://www.ise.rutgers.edu/" rel="noopener noreferrer" target="_blank"> industrial and systems engineering department</a> from 2007 to 2013.“If my journey holds one lesson,” he says, “it is this: Struggle builds resilience, and resilience makes the extraordinary possible. Even in darkness, perseverance lights the way.”

Jan 20, 2026

The Quest to Build a Radio Telescope That Can Hear the Cosmic Dark Ages

<img src="https://spectrum.ieee.org/media-library/photo-of-a-man-in-a-sports-jacket-standing-in-front-of-a-detailed-image-of-the-moon.jpg?id=62815043&width=1245&height=700&coordinates=0%2C559%2C0%2C559"/> Isolation dictates where we go to see into the far reaches of the universe. The <a href="https://spectrum.ieee.org/vera-rubin-observatory-first-images" target="_self">Atacama Desert</a> of Chile, the summit of <a href="https://spectrum.ieee.org/jupiter-auroras-jaxa-nasa" target="_self">Mauna Kea</a> in Hawaii, the vast expanse of the <a href="https://spectrum.ieee.org/far-from-radio-interference-the-square-kilometre-array-takes-root-in-south-africa-and-the-australian-outback" target="_self">Australian Outback</a>—these are where astronomers and engineers have built the great observatories and radio telescopes of modern times. The skies are usually clear, the air is arid, and the electronic din of civilization is far away.It was to one of these places, in the high desert of New Mexico, that a young astronomer named <a href="https://www.colorado.edu/faculty/burns/" target="_blank">Jack Burns</a> went to study radio jets and quasars far beyond the Milky Way. It was 1979, he was just out of grad school, and the <a href="https://public.nrao.edu/telescopes/vla/" target="_blank">Very Large Array</a>, a constellation of 28 giant dish antennas on an open plain, was a new mecca of radio astronomy.But the VLA had its limitations—namely, that Earth’s protective atmosphere and ionosphere blocked many parts of the electromagnetic spectrum, and that, even in a remote desert, earthly interference was never completely gone.Could there be a better, even lonelier place to put a radio telescope? Sure, a NASA planetary scientist named <a href="https://historycollection.jsc.nasa.gov/JSCHistoryPortal/history/oral_histories/NASA_HQ/SMD/MendellWW/MendellWW_9-20-17.htm" target="_blank">Wendell Mendell</a>, told Burns: How about the moon? He asked if Burns had ever thought about building one there.“My immediate reaction was no. Maybe even hell, no. Why would I want to do that?” Burns recalls with a self-deprecating smile. His work at the VLA had gone well, he was fascinated by cosmology’s big questions, and he didn’t want to be slowed by the bureaucratic slog of getting funding to launch a new piece of hardware.But Mendell suggested he do some research and speak at a conference on future lunar observatories, and Burns’s thinking about a space-based radio telescope began to shift. That was in 1984. In the four decades since, he’s published more than <a href="https://ui.adsabs.harvard.edu/search/filter_database_fq_database=OR&filter_database_fq_database=database%3A%22astronomy%22&format=SHORT&fq=%7B!type%3Daqp%20v%3D%24fq_database%7D&fq_database=(database%3A%22astronomy%22)&q=%3Dauthor%3A(%22burns%2C%20jack%22%20OR%20%22burns%2C%20jack%20o.%22%20OR%20%22burns%2C%20j.o.%22)&sort=score%20desc%2C%20bibcode%20desc&unprocessed_parameter=qform&unprocessed_parameter=Require%20Field%20for%20Selection&p_=0" target="_blank">500 peer-reviewed papers</a> on radio astronomy. He’s been an <a href="https://www.nasa.gov/nac/nac-meetings-archive-2009-2016/" target="_blank">adviser</a> to <a href="https://www.colorado.edu/faculty/burns/about-jack" target="_blank">NASA</a>, the Department of Energy, and the White House, as well as a professor and a university administrator. And while doing all that, Burns has had an ongoing second job of sorts, as a quietly persistent advocate for radio astronomy from space.And early next year, if all goes well, a radio telescope for which he’s a scientific investigator will be launched—not just into space, not just to the moon, but to the moon’s far side, where it will observe things invisible from Earth.“You can see we don’t lack for ambition after all these years,” says Burns, now 73 and a professor emeritus of astrophysics at <a href="https://www.colorado.edu/faculty/burns/about-jack" target="_blank">the University of Colorado Boulder</a>.The instrument is called <a href="https://arxiv.org/abs/2301.10345" target="_blank">LuSEE-Night</a>, short for <a href="https://www.bnl.gov/newsroom/news.php?a=122408" target="_blank">Lunar Surface Electromagnetics Experiment–Night</a><a href="https://assets.science.nasa.gov/dynamicimage/assets/science/psd/lunar-science/lunar-landers/Firefly_Blue%20Ghost%20Mission%202_2025.jpg?w=2689&h=1513&fit=clip&crop=faces%2Cfocalpoint" target="_blank">.</a> It will be launched from Florida aboard a SpaceX rocket and carried to the moon’s far side atop a squat four-legged robotic spacecraft called <a href="https://fireflyspace.com/missions/blue-ghost-mission-2/" target="_blank">Blue Ghost Mission 2</a>, built and operated by <a href="https://fireflyspace.com/" target="_blank">Firefly Aerospace</a> of Cedar Park, Texas. <img alt="Illustration of a four-legged structure with solar panels on the sides on the surface of the moon. " class="rm-shortcode" data-rm-shortcode-id="8ac04a163bc8b1148f2abde76401f793" data-rm-shortcode-name="rebelmouse-image" id="f6e66" loading="lazy" src="https://spectrum.ieee.org/media-library/illustration-of-a-four-legged-structure-with-solar-panels-on-the-sides-on-the-surface-of-the-moon.png?id=62815356&width=980"/>In an artist’s rendering, the LuSEE-Night radio telescope sits atop Firefly Aerospace’s Blue Ghost 2 lander, which will carry it to the moon’s far side. Firefly Aerospace Landing will be <a href="https://spectrum.ieee.org/lunar-landing" target="_self">risky</a>: Blue Ghost 2 will be on its own, in a place that’s out of the sight of ground controllers. But Firefly’s <a href="https://fireflyspace.com/missions/blue-ghost-mission-1/" target="_blank">Blue Ghost 1</a> pulled off the first successful landing by a private company on the moon’s near side in March 2025. And Burns has already put hardware on the lunar surface, albeit with mixed results: An experiment he helped conceive was on board a lander called <a href="https://www.intuitivemachines.com/im-1" target="_blank">Odysseus</a>, built by Houston-based <a href="https://www.intuitivemachines.com/" target="_blank">Intuitive Machines</a>, in 2024. Odysseus was damaged on landing, but Burns’s experiment still returned some useful data.Burns says he’d be bummed about that 2024 mission if there weren’t so many more coming up. He’s joined in proposing myriad designs for radio telescopes that could go to the moon. And he’s kept going through political disputes, technical delays, even a confrontation with cancer. Finally, finally, the effort is paying off.“We’re getting our feet into the lunar soil,” says Burns, “and understanding what is possible with these radio telescopes in a place where we’ve never observed before.”<h2>Why Go to the Far Side of the Moon? </h2>A moon-based radio telescope could help unravel some of the greatest mysteries in space science. Dark matter, dark energy, neutron stars, and gravitational waves could all come into better focus if observed from the moon. One of Burns’s collaborators on LuSEE-Night, astronomer <a href="https://pma.caltech.edu/people/gregg-w-hallinan" target="_blank">Gregg Hallinan</a> of Caltech, would like such a telescope to further his research on <a href="https://www.kiss.caltech.edu/impact/Hallinan.html" target="_blank">electromagnetic activity around exoplanets</a>, a possible measure of whether these distant worlds are habitable. Burns himself is especially interested in the <a href="https://www.hilarispublisher.com/open-access/physics-of-cosmic-dark-ages-100436.html" target="_blank">cosmic dark ages</a>, an epoch that began more than 13 billion years ago, just 380,000 years after the big bang. The young universe had cooled enough for neutral hydrogen atoms to form, which trapped the light of stars and galaxies. The dark ages lasted between 200 million and 400 million years.<h3></h3> <div class="flourish-embed flourish-timeline" data-src="visualisation/27231096?602891"><script src="https://public.flourish.studio/resources/embed.js"></script><noscript><img alt="timeline visualization" src="https://public.flourish.studio/visualisation/27231096/thumbnail" width="100%"/></noscript></div>LuSEE-Night will listen for faint signals from the cosmic dark ages, a period that began about 380,000 years after the big bang, when neutral hydrogen atoms had begun to form, trapping the light of stars and galaxies. Chris Philpot“It’s a critical period in the history of the universe,” says Burns. “But we have no data from it.”The problem is that residual radio signals from this epoch are very faint and easily drowned out by closer noise—in particular, our earthly communications networks, power grids, radar, and so forth. The sun adds its share, too. What’s more, these early signals have been dramatically redshifted by the expansion of the universe, their wavelengths stretched as their sources have sped away from us over billions of years. The most critical example is neutral hydrogen, the most abundant element in the universe, which when excited in the laboratory emits a radio signal with a wavelength of 21 centimeters. Indeed, with just <a href="https://spectrum.ieee.org/software-defined-radio" target="_self">some backyard equipment</a>, you can easily detect neutral hydrogen in nearby galactic gas clouds close to that wavelength, which corresponds to a frequency of 1.42 gigahertz. But if the hydrogen signal originates from the dark ages, those 21 centimeters are lengthened to tens of meters. That means scientists need to listen to frequencies well below 50 megahertz—parts of the radio spectrum that are largely blocked by Earth’s ionosphere.Which is why the lunar far side holds such appeal. It may just be the quietest site in the inner solar system.“It really is the only place in the solar system that never faces the Earth,” says <a href="https://astro.berkeley.edu/people/david-deboer/" target="_blank">David DeBoer</a>, a research astronomer at the University of California, Berkeley. “It really is kind of a wonderful, unique place.”For radio astronomy, things get even better during the lunar night, when the sun drops beneath the horizon and is blocked by the moon’s mass. For up to 14 Earth-days at a time, a spot on the moon’s far side is about as electromagnetically dark as any place in the inner solar system can be. No radiation from the sun, no confounding signals from Earth. There may be signals from a few distant space probes, but otherwise, ideally, your antenna only hears the raw noise of the cosmos.“When you get down to those very low radio frequencies, there’s a source of noise that appears that’s associated with the solar wind,” says Caltech’s Hallinan. Solar wind is the stream of charged particles that speed relentlessly from the sun. “And the only location where you can escape that within a billion kilometers of the Earth is on the lunar surface, on the nighttime side. The solar wind screams past it, and you get a cavity where you can hide away from that noise.”<h2>How Does LuSEE-Night Work? </h2>LuSEE-Night’s receiver looks simple, though there’s really nothing simple about it. Up top are <a href="https://arxiv.org/html/2407.07173v1#S2" target="_blank">two dipole antennas</a>, each of which consists of two collapsible rods pointing in opposite directions. The dipole antennas are mounted perpendicular to each other on a small <a href="https://physicalsciences.lbl.gov/2025/09/15/berkeley-lab-engineer-designs-hardware-for-the-moons-unforgiving-environment/" target="_blank">turntable</a>, forming an X when seen from above. Each dipole antenna extends to about 6 meters. The turntable sits atop a box of support equipment that’s a bit less than a cubic meter in volume; the equipment bay, in turn, sits atop the <a href="https://science.nasa.gov/lunar-science/clps-deliveries/cs-3/" target="_blank">Blue Ghost 2</a> lander, a boxy spacecraft about 2 meters tall. <img alt="A person wearing a hairnet, facemask, and vinyl gloves working on a shiny metal apparatus." class="rm-shortcode" data-rm-shortcode-id="aca2f5aef83aa4834d9a616a5e9be5ed" data-rm-shortcode-name="rebelmouse-image" id="3e129" loading="lazy" src="https://spectrum.ieee.org/media-library/a-person-wearing-a-hairnet-facemask-and-vinyl-gloves-working-on-a-shiny-metal-apparatus.png?id=62822192&width=980"/> <img alt="A photo of people wearing hairnets, facemasks, and vinyl gloves working on a shiny metal apparatus. " class="rm-shortcode" data-rm-shortcode-id="28e2b518fcdd365420b2780667e05259" data-rm-shortcode-name="rebelmouse-image" id="341a7" loading="lazy" src="https://spectrum.ieee.org/media-library/a-photo-of-people-wearing-hairnets-facemasks-and-vinyl-gloves-working-on-a-shiny-metal-apparatus.png?id=62822190&width=980"/> <img alt="A person wearing a hairnet, facemask, and vinyl gloves working on a shiny metal apparatus." class="rm-shortcode" data-rm-shortcode-id="ff7c51697658f771e1dfb402ed4492cf" data-rm-shortcode-name="rebelmouse-image" id="76e4f" loading="lazy" src="https://spectrum.ieee.org/media-library/a-person-wearing-a-hairnet-facemask-and-vinyl-gloves-working-on-a-shiny-metal-apparatus.png?id=62822146&width=980"/> LuSEE-Night undergoes final assembly [top and center] at the Space Sciences Laboratory at the University of California, Berkeley, and testing [bottom] at Firefly Aerospace outside Austin, Texas. From top: Space Sciences Laboratory/University of California, Berkeley (2); Firefly Aerospace “It’s a beautiful instrument,” says <a href="https://physics.berkeley.edu/people/faculty/stuart-bale" target="_blank">Stuart Bale</a>, a physicist at the University of California, Berkeley, who is NASA’s principal investigator for the project. “We don’t even know what the radio sky looks like at these frequencies without the sun in the sky. I think that’s what LuSEE-Night will give us.”The apparatus was designed to serve several incompatible needs: It had to be sensitive enough to detect very weak signals from deep space; rugged enough to withstand the extremes of the lunar environment; and quiet enough to not interfere with its own observations, yet loud enough to talk to Earth via relay satellite as needed. Plus the instrument had to stick to a budget of about US $40 million and not weigh more than 120 kilograms. The mission plan calls for two years of operations.The antennas are made of a beryllium copper alloy, chosen for its high conductivity and stability as lunar temperatures plummet or soar by as much as 250 °C every time the sun rises or sets. LuSEE-Night will make precise voltage measurements of the signals it receives, using a high-impedance junction field-effect transistor to act as an amplifier for each antenna. The signals are then fed into a spectrometer—the main science instrument—which reads those voltages at <a href="https://arxiv.org/pdf/2301.10345" target="_blank">102.4 million samples per second</a>. That high read-rate is meant to prevent the exaggeration of any errors as faint signals are amplified. Scientists believe that a cosmic dark-ages signature would be five to six orders of magnitude weaker than the other signals that LuSEE-Night will record.The turntable is there to help characterize the signals the antennas receive, so that, among other things, an ancient dark-ages signature can be distinguished from closer, newer signals from, say, galaxies or interstellar gas clouds. Data from the early universe should be virtually <a href="https://adsabs.harvard.edu/full/1973ApJ...180..317C%22%3ECollins" target="_blank">isotropic</a>, meaning that it comes from all over the sky, regardless of the antennas’ orientation. Newer signals are more likely to come from a specific direction. Hence the turntable: If you collect data over the course of a lunar night, then reorient the antennas and listen again, you’ll be better able to distinguish the distant from the very, very distant.What’s the ideal lunar <a href="https://arxiv.org/html/2407.07173v1#S3" target="_blank">landing spot</a> if you want to take such readings? One as nearly opposite Earth as possible, on a flat plain. Not an easy thing to find on the moon’s hummocky far side, but mission planners pored over maps made by lunar satellites and chose a prime location about 24 degrees south of the lunar equator.Other lunar telescopes have been proposed for placement in the permanently shadowed craters near the <a href="https://spectrum.ieee.org/moon-base" target="_self">moon’s south pole</a>, just over the horizon when viewed from Earth. Such craters are coveted for the water ice they may hold, and the low temperatures in them (below -240 °C) are great if you’re doing infrared astronomy and need to keep your instruments cold. But the location is terrible if you’re working in long-wavelength radio.“Even the inside of such craters would be hard to shield from Earth-based radio frequency interference (RFI) signals,” <a href="https://leonkoopmans.com/about/" target="_blank">Leon Koopmans</a> of the University of Groningen in the Netherlands, said in an email. “They refract off the crater rims and often, due to their long wavelength, simply penetrate right through the crater rim.”RFI is a major—and sometimes maddening—issue for sensitive instruments. The first-ever landing on the lunar far side was by the Chinese <a href="https://www.cnsa.gov.cn/english/n6465652/n6465653/c6805233/content.html" target="_blank">Chang’e 4</a> spacecraft, in 2019. It carried a <a href="https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2022RS007595" target="_blank">low-frequency radio spectrometer</a>, among other experiments. But it failed to return meaningful results, Chinese researchers said, mostly because of interference from the spacecraft itself.<h2>The Accidental Birth of Radio Astronomy </h2>Sometimes, though, a little interference makes history. Here, it’s worth a pause to remember <a href="https://legacy.nrao.edu/epo/aoc/puente/jansky/hist_jansky.shtml" target="_blank">Karl Jansky</a>, considered the <a href="https://aas.org/posts/news/2018/07/month-astronomical-history-1" target="_blank">father of radio astronomy</a>. In 1928, he was a young engineer at Bell Telephone Laboratories in Holmdel, N.J., assigned to isolate <a href="https://www.aps.org/archives/publications/apsnews/201505/physicshistory.cfm" target="_blank">sources of static in shortwave transatlantic telephone calls</a>. Two years later, he built a <a href="https://www.bigear.org/CSMO/HTML/CS12/cs12p08.htm" target="_blank">30-meter-long directional antenna</a>, mostly out of brass and wood, and after accounting for thunderstorms and the like, there was still noise he couldn’t explain. At first, its strength seemed to follow a daily cycle, rising and sinking with the sun. But after a few months’ observation, the sun and the noise were badly out of sync. <img alt="Black and white photo of a man standing in a field in front of a large structure made of crisscrossing segments and resting on wheels. " class="rm-shortcode" data-rm-shortcode-id="61aae093ae07d953a2386e9caccb4ac0" data-rm-shortcode-name="rebelmouse-image" id="8d525" loading="lazy" src="https://spectrum.ieee.org/media-library/black-and-white-photo-of-a-man-standing-in-a-field-in-front-of-a-large-structure-made-of-crisscrossing-segments-and-resting-on-w.png?id=62821778&width=980"/> In 1930, Karl Jansky, a Bell Labs engineer in Holmdel, N.J., built this rotating antenna on wheels to identify sources of static for radio communications. NRAO/AUI/NSF It gradually became clear that the noise’s period wasn’t 24 hours; it was 23 hours and 56 minutes—the time it takes Earth to turn once relative to the stars. The strongest interference seemed to come from the direction of the constellation Sagittarius, which optical astronomy suggested was the center of the Milky Way. In 1933, Jansky published a paper in <a href="https://proceedingsoftheieee.ieee.org/about/history/" target="_blank">Proceedings of the Institute of Radio Engineers</a> with a provocative title: “<a href="https://doi.org/10.1109/JPROC.1998.681378" target="_blank">Electrical Disturbances Apparently of Extraterrestrial Origin</a>.” He had opened the electromagnetic spectrum up to astronomers, even though he never got to pursue radio astronomy himself. The interference he had defined was, to him, “star noise.”Thirty-two years later, two other Bell Labs scientists, <a href="https://www.nobelprize.org/prizes/physics/1978/penzias/biographical/" target="_blank">Arno Penzias</a> and <a href="https://www.nobelprize.org/prizes/physics/1978/wilson/facts/" target="_blank">Robert Wilson</a>, ran into some interference of their own. In 1965 they were trying to adapt a <a href="https://npgallery.nps.gov/NRHP/GetAsset/NHLS/89002457_text" target="_blank">horn antenna</a> in Holmdel for radio astronomy—but there was a hiss, in the <a href="https://aether.lbl.gov/www/science/cmb.html" target="_blank">microwave</a> band, coming from all parts of the sky. They had no idea what it was. They ruled out interference from New York City, not far to the north. They rewired the receiver. They cleaned out bird droppings in the antenna. Nothing worked. <img alt="Black and white photo of a large triangular structure on a frame, with two people looking up at it. " class="rm-shortcode" data-rm-shortcode-id="4b1d9f82c9bdc76a35cdd377c2809677" data-rm-shortcode-name="rebelmouse-image" id="fa4de" loading="lazy" src="https://spectrum.ieee.org/media-library/black-and-white-photo-of-a-large-triangular-structure-on-a-frame-with-two-people-looking-up-at-it.jpg?id=62822141&width=980"/> In the 1960s, Arno Penzias and Robert W. Wilson used this horn antenna in Holmdel, N.J., to detect faint signals from the big bang. GL Archive/Alamy Meanwhile, an hour’s drive away, a team of physicists at Princeton University under <a href="https://phy.princeton.edu/department/history/faculty-history/robert-dicke" target="_blank">Robert Dicke</a> was trying to find proof of the big bang that began the universe 13.8 billion years ago. They theorized that it would have left a hiss, in the microwave band, coming from all parts of the sky. They’d begun to build an antenna. Then Dicke got a phone call from Penzias and Wilson, looking for help. “Well, boys, we’ve been scooped,” he famously said when the call was over. Penzias and Wilson had accidentally found the <a href="https://spectrum.ieee.org/big-bang-theory-discovery" target="_self">cosmic microwave background</a>, or CMB, the leftover radiation from the big bang.Burns and his colleagues are figurative heirs to Jansky, Penzias, and Wilson. Researchers suggest that the giveaway signature of the cosmic dark ages may be a <a href="https://www.bnl.gov/newsroom/news.php?a=221439" target="_blank">minuscule dip</a> in the CMB. They theorize that dark-ages hydrogen may be detectable only because it has been absorbing a little bit of the microwave energy from the dawn of the universe.<h2>The Moon Is a Harsh Mistress </h2>The plan for Blue Ghost Mission 2 is to touch down soon after the sun has risen at the landing site. That will give mission managers two weeks to check out the spacecraft, take pictures, conduct other experiments that Blue Ghost carries, and charge LuSEE-Night’s battery pack with its photovoltaic panels. Then, as local sunset comes, they’ll turn everything off except for the LuSEE-Night receiver and a bare minimum of support systems. <img alt="Image of the moon's surface, with a closeup of one section. " class="rm-shortcode" data-rm-shortcode-id="bc6fd1e96f6351b1efe30ebdfd0ec1f5" data-rm-shortcode-name="rebelmouse-image" id="5b48b" loading="lazy" src="https://spectrum.ieee.org/media-library/image-of-the-moon-s-surface-with-a-closeup-of-one-section.png?id=62815382&width=980"/>LuSEE-Night will land at a site [orange dot] that’s about 25 degrees south of the moon’s equator and opposite the center of the moon’s face as seen from Earth. The moon’s far side is ideal for radio astronomy because it’s shielded from the solar wind as well as signals from Earth. Arizona State University/GSFC/NASA There, in the frozen electromagnetic stillness, it will scan the spectrum between 0.1 and 50 MHz, gathering data for a low-frequency <a href="https://arxiv.org/html/2508.16773v1" target="_blank">map</a> of the sky—maybe including the first tantalizing signature of the dark ages.“It’s going to be really tough with that instrument,” says Burns. “But we have some hardware and software techniques that…we’re hoping will allow us to detect what’s called the global or all-sky signal.… We, in principle, have the sensitivity.” They’ll listen and listen again over the course of the mission. That is, if their equipment doesn’t freeze or fry first.A major task for LuSEE-Night is to protect the <a href="https://arxiv.org/html/2407.07173v1" target="_blank">electronics</a> that run it. <a href="https://science.nasa.gov/moon/weather-on-the-moon/" target="_blank">Temperature extremes</a> are the biggest problem. Systems can be hardened against cosmic radiation, and a sturdy spacecraft should be able to handle the stresses of launch, flight, and landing. But how do you build it to last when temperatures range between 120 and −130 °C? With layers of insulation? Electric heaters to reduce nighttime chill?“All of the above,” says Burns. To reject daytime heat, there will be a multicell parabolic radiator panel on the outside of the equipment bay. To keep warm at night, there will be battery power—a lot of battery power. Of LuSEE-Night’s launch mass of 108 kg, about 38 kg is a lithium-ion battery pack with a capacity of 7,160 watt-hours, mostly to generate heat. The battery cells will recharge photovoltaically after the sun rises. The all-important spectrometer has been programmed to cycle off periodically during the two weeks of darkness, so that the battery’s state of charge doesn’t drop below 8 percent; better to lose some observing time than lose the entire apparatus and not be able to revive it.<h2>Lunar Radio Astronomy for the Long Haul </h2>And if they can’t revive it? Burns has been through that before. In 2024 he watched helplessly as Odysseus, the first U.S.-made lunar lander in 50 years, touched down—and then went silent for 15 agonizing minutes until controllers in Texas realized they were receiving only occasional pings instead of detailed data. Odysseus <a href="https://spectrum.ieee.org/lunar-landing-intuitive-machines" target="_self">had landed</a> hard, snapped a leg, and ended up lying almost on its side. <img alt="Color photo of a metal structure inside an open rocket. " class="rm-shortcode" data-rm-shortcode-id="09fd5a25e9cdd1ac73cdfb825814bb37" data-rm-shortcode-name="rebelmouse-image" id="97286" loading="lazy" src="https://spectrum.ieee.org/media-library/color-photo-of-a-metal-structure-inside-an-open-rocket.png?id=62822546&width=980"/> ROLSES-1, shown here inside a SpaceX Falcon 9 rocket, was the first radio telescope to land on the moon, in February 2024. During a hard landing, one leg broke, making it difficult for the telescope to send readings back to Earth.Intuitive Machines/SpaceXAs part of its scientific cargo, Odysseus carried <a href="https://arxiv.org/html/2503.09842v2#S3" target="_blank">ROLSES-1</a> (Radiowave Observations on the Lunar Surface of the photo-Electron Sheath), an experiment Burns and a friend had suggested to NASA years before. It was partly a test of technology, partly to study the complex interactions between sunlight, radiation, and lunar soil—there’s enough electric charge in the soil sometimes that dust particles levitate above the moon’s surface, which could potentially mess with radio observations. But Odysseus was damaged badly enough that instead of a week’s worth of data, ROLSES got 2 hours, most of it recorded before the landing. A grad student working with Burns, <a href="https://www.researchgate.net/scientific-contributions/Joshua-J-Hibbard-2182593048" target="_blank">Joshua Hibbard</a>, managed to partially salvage the experiment and prove that ROLSES had worked: Hidden in its raw data were <a href="https://iopscience.iop.org/article/10.3847/1538-3881/ae18d8" target="_blank">signals from Earth and the Milky Way</a>.“It was a harrowing experience,” Burns said afterward, “and I’ve told my students and friends that I don’t want to be first on a lander again. I want to be second, so that we have a greater chance to be successful.” He says he feels good about LuSEE-Night being on the Blue Ghost 2 mission, especially after the successful Blue Ghost 1 landing. The ROLSES experiment, meanwhile, will get a second chance: ROLSES-2 has been scheduled to fly on <a href="https://fireflyspace.com/missions/blue-ghost-mission-3/" target="_blank">Blue Ghost Mission 3</a>, perhaps in 2028. <img alt="Artist\u2019s rendering of a gray surface with parallel zigzagging lines. " class="rm-shortcode" data-rm-shortcode-id="99103eff7bd449b73b15a8dd9489b922" data-rm-shortcode-name="rebelmouse-image" id="bbc57" loading="lazy" src="https://spectrum.ieee.org/media-library/artist-u2019s-rendering-of-a-gray-surface-with-parallel-zigzagging-lines.png?id=62821827&width=980"/> NASA’s plan for the FarView Observatory lunar radio telescope array, shown in an artist’s rendering, calls for 100,000 dipole antennas to be spread out over 200 square kilometers. Ronald Polidan If LuSEE-Night succeeds, it will doubtless raise questions that require much <a href="https://arxiv.org/abs/2504.03418" target="_blank">more ambitious radio telescopes</a>. Burns, Hallinan, and others have already gotten early NASA funding for a giant interferometric array on the moon called <a href="https://www.sciencedirect.com/science/article/pii/S0273117724003405?via%3Dihub" target="_blank">FarView</a>. It would consist of a grid of 100,000 antenna nodes spread over 200 square kilometers, made of aluminum <a href="https://spectrum.ieee.org/blue-origin-molten-regolith-electrolysis" target="_self">extracted from lunar soil</a>. They say assembly could begin as soon as the 2030s, although political and budget realities may get in the way.Through it all, Burns has gently pushed and prodded and lobbied, advocating for a lunar observatory through the terms of ten NASA administrators and seven U.S. presidents. He’s probably learned more about Washington politics than he ever wanted. American presidents have a habit of reversing the space priorities of their predecessors, so missions have sometimes proceeded full force, then languished for years. With LuSEE-Night finally headed for launch, Burns at times sounds buoyant: “Just think. We’re actually going to do cosmology from the moon.” At other times, he’s been blunt: “I never thought—none of us thought—that it would take 40 years.”“Like anything in science, there’s no guarantee,” says Burns. “But we need to look.”

Jan 18, 2026

NASA Demolishes Historic Test Stands That Built the Space Age

<img src="https://spectrum.ieee.org/media-library/a-spacecraft-being-lowered-by-crane-from-a-nasa-space-center-test-stand.jpg?id=62826872&width=1245&height=700&coordinates=0%2C228%2C0%2C229"/> The thunderous roar that echoed across Huntsville, Alabama, on 10 January wasn’t a rocket launch but something equally momentous: the end of an era. Two massive test stands at <a href="https://www.nasa.gov/marshall/" target="_blank">Marshall Space Flight Center</a> that helped send humans to the moon <a href="https://www.nasa.gov/centers-and-facilities/marshall/nasa-marshall-removes-2-historic-test-stands/" target="_blank">collapsed in carefully choreographed implosions</a>, their steel frameworks crumbling in seconds after decades standing as monuments to U.S. spaceflight achievement.The <a href="https://www.nps.gov/articles/saturn-v-dynamic-test-stand.htm" target="_blank">Dynamic Test Stand</a> and the <a href="https://www.nps.gov/articles/propulsion-and-structural-test-facility.htm" target="_blank">Propulsion and Structural Test Facility</a>, better known as the T-tower for its distinctive shape, represented more than just obsolete infrastructure. Built in the 1950s and ’60s, these structures witnessed the birth of the space age, serving as proving grounds where engineers pushed the limits of rocket technology and ensured every component could withstand the violence of launch.<div class="badge_module shortcode-media shortcode-media-rebelmouse-image rm-float-left rm-resized-container rm-resized-container-25"> <a class="rm-stats-tracked" href="https://www.universetoday.com/" target="_blank"> <img alt='Universe Today logo; text reads "This post originally appeared on Universe Today."' class="rm-shortcode rm-lazyloadable-image" src="https://spectrum.ieee.org/media-library/universe-today-logo-text-reads-this-post-originally-appeared-on-universe-today.png?id=60568425&width=1800&quality=85"/></a> </div> <h2>T-tower’s Role in Rocket Testing</h2>The T-tower came first, constructed in 1957 by the <a href="https://history.redstone.army.mil/space-abma.html" target="_blank">Army Ballistic Missile Agency</a> before NASA even existed. At just over 50 meters tall, it was designed for static testing, where rockets are fired at full power while restrained and connected to instruments that measure every vibration, temperature spike, and pressure fluctuation. Here, engineers tested components of the Saturn family of launch vehicles under the direction of Wernher von Braun, including the mighty F-1 engines that would eventually power <a data-linked-post="2650248198" href="https://spectrum.ieee.org/apollo-13-we-have-a-solution" target="_blank">Apollo missions</a>. The tower later proved essential for testing space shuttle solid rocket boosters before being retired in the 1990s.The Dynamic Test Stand told an even more dramatic story. Built in 1964 and rising over 105 meters above the Alabama landscape, it once stood as the tallest human-made structure in North Alabama. Unlike the T-tower’s static tests, this facility subjected fully assembled Saturn V rockets to the mechanical stresses and vibrations they would experience during actual flight, everything shaking, flexing, and straining just as it would during launch, but without leaving the ground. Engineers couldn’t afford failures once these rockets reached the launchpad at Kennedy Space Center: Saturn V was too powerful, too expensive, and too important to risk.The stand’s role didn’t end with Apollo. In 1978, it became the first location where engineers integrated all space shuttle elements together: orbiter, external fuel tank, and solid rocket boosters assembled as one complete system. Its final mission came in the early 2000s, when it served as a drop tower for microgravity experiments, a far quieter purpose than its explosive origins.Both facilities earned designations as National Historic Landmarks in 1985, recognition of their irreplaceable contributions to human spaceflight. That makes their demolition bittersweet but necessary. The structures are no longer safe, and maintaining aging facilities drains resources that could support current missions. Marshall is removing 19 obsolete structures as part of a broader campus transformation, creating a modern, interconnected facility ready for NASA’s next chapter.“These facilities helped NASA make history. While it is hard to let them go, they’ve earned their retirement. The people who built and managed these facilities and empowered our mission of space exploration are the most important part of their legacy,” said acting Marshall director Rae Ann Meyer in a statement.NASA has worked to preserve that legacy. Detailed architectural drawings, photographs, and written histories now reside permanently in the Library of Congress. Auburn University created high-resolution digital models using <a data-linked-post="2665763899" href="https://spectrum.ieee.org/lidar" target="_blank">LiDAR</a> and 360-degree photography, capturing the structures in exquisite detail before their destruction. These virtual archives ensure future generations can still appreciate the scale and engineering achievement these towers represented, even after the steel has been cleared away.

Jan 17, 2026

Are There Enough Engineers for the AI Boom?

<img src="https://spectrum.ieee.org/media-library/aero-gas-turbines-being-installed-in-a-desert-like-landscape.jpg?id=62826721&width=1245&height=700&coordinates=0%2C187%2C0%2C188"/> <a rel="noopener noreferrer" target="_blank">The AI data center construction boom continues unabated, with the demand for power in the United States potentially reaching </a><a href="https://about.bnef.com/insights/clean-energy/ai-and-the-power-grid-where-the-rubber-meets-the-road/" rel="noopener noreferrer" target="_blank">106 gigawatts by 2035</a>, according to a December report from research and analysis company BloombergNEF. That’s a 36 percent jump from the company’s previous outlook, published just seven months earlier. But there are severe constraints in power availability, material, equipment, and—perhaps most significantly—a lack of engineers, technicians, and skilled craftsmen that could turn the data center boom into a bust.<a href="#_msocom_1" target="_blank"></a>The power grid engineering workforce is <a href="https://spectrum.ieee.org/power-engineering-workforce-gap" target="_self">currently shrinking</a>, and data center operators are also hurting for trained <a href="https://spectrum.ieee.org/data-center-jobs" target="_self">electrical engineers</a>. Laura Laltrello, the chief operating officer for <a href="https://www.applieddigital.com/" target="_blank">Applied Digital</a>, says demand has accelerated for civil, mechanical, and electrical engineers, as well as construction management and oversight positions in recent months. (Applied Digital is a data center developer and operator that is building two data center campuses near Harwood, North Dakota, that will require 1.4 GW of power when completed.) The growing demand for skilled workers has forced her company to widen the recruitment perimeter.“As we anticipate a shortage of traditional engineering talent, we are sourcing from diverse industries,” says Laltrello. “We are finding experts who understand power and cooling from sectors like nuclear energy, the military, and aerospace. Expertise doesn’t have to come from a data center background.”<h2>Growing Demand for Data Center Engineers</h2>For every engineer needed to design, specify, build, inspect, commission, or run a new AI data center, dozens of other positions are in short supply. According to the Association for Computer Operations and Management’s (AFCOM) <a href="https://afcom.com/news/news.asp?id=692525" target="_blank">State of the Data Center Report 2025</a>, 58 percent of data center managers identified multiskilled data center operators as the top area of growth, while 50 percent signaled increasing demand for data center engineers. Security specialists are also a critical need.Through the next decade, the U.S. Bureau of Labor Statistics projects the need for almost <a href="https://www.bls.gov/opub/btn/volume-14/whats-behind-the-projected-construction-employment-growth-from-2023-to-2033.htm" target="_blank">400,000 more construction workers by 2033</a>. By far the biggest needs are in power infrastructure, electricians, plumbing, and HVAC, <a target="_blank"></a><a rel="noopener noreferrer" target="_blank">and roughly </a><a href="https://www.bls.gov/ooh/architecture-and-engineering/electrical-and-electronics-engineers.htm" rel="noopener noreferrer" target="_blank">17,500 electrical and electronics engineers</a>. These categories directly map to the skills required to design, build, commission, and operate modern data centers.“The challenge is not simply the absolute number of workers available, but the timing and intensity of demand,” says Bill Kleyman, author of the AFCOM report and the CEO of AI infrastructure firm <a href="https://www.apolo.us/" target="_blank">Apolo</a>. “Data centers are expanding at the same time that utilities, manufacturing, renewables, grid infrastructure, and construction are all competing for the same skilled labor pool, and AI is amplifying this pressure.”Data center developers like <a href="https://lancium.com/" target="_blank">Lancium</a> and construction firms like <a href="https://www.crusoe.ai/" rel="noopener noreferrer" target="_blank">Crusoe</a> face enormous demands to build faster, bigger, and more power-dense facilities. For example, they’re developing the Stargate project in Abilene, Texas, for Oracle and OpenAI. The project has two buildings that went live in October 2025, with another six scheduled for completion by the middle of 2026. The entire AI data center campus, once completed, will require 1.2 GW of power.<a href="#_msocom_6" target="_blank"></a><a href="https://www.linkedin.com/in/michael-mcnamara-1055211/" target="_blank">Michael McNamara</a>, the CEO of <a href="https://lancium.com/" target="_blank">Lancium</a>, says that in one year his company can currently build enough AI data center infrastructure to require 1 GW of power. Big tech firms, he says, want this raised to 1 GW a quarter and eventually 1 GW per month or less.That kind of ramp-up of construction pace calls for tens of thousands more engineers. The shortage of engineering talent is paralleled by persistent staffing shortages in data center operations and facility management professionals, electrical and mechanical technicians, high-voltage and power systems engineers, skilled HVAC technicians with experience in high-density or liquid cooling, and construction specialists familiar with complex mechanical, electrical, and plumbing (MEP) integration, says <a href="https://www.linkedin.com/in/matthewhawkins/" rel="noopener noreferrer" target="_blank">Matthew Hawkins</a>, the <a rel="noopener noreferrer" target="_blank"></a>director of education for Uptime Institute.“Demand for each category is rising significantly faster than supply,” says Hawkins.<a rel="noopener noreferrer" target="_blank"></a><a rel="noopener noreferrer" target="_blank"></a><a rel="noopener noreferrer" target="_blank"></a>Technical colleges and applied education programs are among the most effective engines for workforce growth in the data center industry. They focus on hands-on skills, facilities operations, power and cooling systems, and real-world job readiness. With so many new data centers being built in Texas, workforce programs are popping up all over that state. One example is the SMU Lyle School of Engineering’s Master of Science in Datacenter Systems Engineering (MS DSE) in Dallas. The program blends electrical engineering, IT, facilities management, business continuity, and cybersecurity. There is also a 12-week AI data center technician program at Dallas College and a similar program at Texas State Technical College near Waco.“Technical colleges are driving the charge in bringing new talent to an industry undergoing exponential growth with an almost infinite appetite for skilled workers,” says Wendy Schuchart, an association manager at AFCOM.Vendors and industry associations are actively addressing the talent gap too. Microsoft’s <a href="https://careers.microsoft.com/v2/global/en/datacenteracademy.html" rel="noopener noreferrer" target="_blank">Datacenter Academy</a> is a public-private partnership involving community colleges in regions where Microsoft operates data center facilities. <a rel="noopener noreferrer" target="_blank"></a><a href="https://datacenters.google/workforce-development-program/" rel="noopener noreferrer" target="_blank">Google supports</a> local nonprofits and colleges offering training in IT and data center operations, and <a href="https://www.amazon.jobs/content/en/teams/amazon-web-services/technical-apprenticeships" rel="noopener noreferrer" target="_blank">Amazon offers data center apprenticeships</a>.The <a href="https://www.siemens.com/us/en/products/energy/low-voltage/siemens-educates-america.html" rel="noopener noreferrer" target="_blank">Siemens Educates America</a> program has surpassed 32,000 apprenticeships across 32 states, 36 labs, and 72 partner industry labor organizations. The company has committed to training 200,000 electricians and electrical manufacturing workers by 2030. Similarly, the National Electrical Contractors Association (NECA) operates the <a href="https://www.electricaltrainingalliance.org/AboutUs" rel="noopener noreferrer" target="_blank">Electrical Training Alliance</a>; the Society of Manufacturing Engineers (SME) offers <a href="https://learn.toolingu.com/" rel="noopener noreferrer" target="_blank">ToolingU-SME</a>, aimed at expanding the manufacturing workforce; and <a href="https://uptimeinstitute.com/education" rel="noopener noreferrer" target="_blank">Uptime Institute Education</a> programs look to accelerate the readiness of technicians and operators.<a href="#_msocom_12" target="_blank"></a>“Every university we speak with is thinking about this challenge and shifting its curriculum to prepare students for the future of digital infrastructure,” said Laltrello. “The best way to predict the future is to build it.”

Jan 16, 2026

IEEE Medal of Honor Recipient Is Nvidia’s CEO Jensen Huang

<img src="https://spectrum.ieee.org/media-library/person-standing-at-a-podium-during-the-ieee-medal-of-honor-ceremony-vibrant-background-lights.jpg?id=62821722&width=1245&height=700&coordinates=0%2C469%2C0%2C469"/> <a href="https://nvidianews.nvidia.com/bios/jensen-huang" rel="noopener noreferrer" target="_blank">Jensen Huang</a>, founder and CEO of <a href="https://www.nvidia.com/en-us/" rel="noopener noreferrer" target="_blank">Nvidia</a>, is the 2026 <a href="https://corporate-awards.ieee.org/ieee-medal-of-honor/" rel="noopener noreferrer" target="_blank">IEEE Medal of Honor</a> recipient. The IEEE honorary member is being recognized for his “leadership in the development of graphics processing units and their application to scientific computing and artificial intelligence.” The news was announced on 6 January by IEEE’s president and CEO, <a href="https://spectrum.ieee.org/mary-ellen-randall-2025-ieee-president-elect" target="_self">Mary Ellen Randall</a>, at the <a href="https://spectrum.ieee.org/ces-2026-preview" target="_self">Consumer Electronics Show</a> in Las Vegas.Huang helped found Nvidia in 1993. Under his direction, the company introduced the programmable GPU six years later. The device sparked extraordinary advancements that have transformed fields including artificial intelligence, computing, and medicine—influencing how technology improves society.“[Receiving the IEEE Medal of Honor] is an incredible honor, ” Huang said at <a href="https://www.youtube.com/watch?v=xj99M127aec" rel="noopener noreferrer" target="_blank">the CES event</a>. “I thank [IEEE] for this incredible award that I receive on behalf of all the great employees at Nvidia.”With a US $2 million prize the award underscores IEEE’s commitment to celebrating visionaries who drive the future of technology for the benefit of humanity.“The IEEE Medal of Honor is the pinnacle of recognition and our most prestigious award,” Randall said <a href="https://www.youtube.com/watch?v=xj99M127aec" rel="noopener noreferrer" target="_blank">at the event</a>. “[Jensen] Huang’s leadership and technical vision have unlocked a new era of innovation.“His vision and subsequent development of [Nvidia’s first GPU hardware] is emblematic of the [award].”<h2>Huang’s impact on technology</h2>Huang’s impact has been acknowledged beyond the realm of engineering. He was named as one of the “Architects of AI,” a group of eight tech leaders who were collectively named <a href="https://time.com/7339685/person-of-the-year-2025-ai-architects/" rel="noopener noreferrer" target="_blank">Time magazine’s 2025 Person of the Year</a>. He was also featured on a 2021 cover of <a href="https://time.com/collection/100-most-influential-people-2021/6095971/jensen-huang/" rel="noopener noreferrer" target="_blank">Time magazine</a>, was named the world’s top-performing CEO for 2019 by <a href="https://hbr.org/" rel="noopener noreferrer" target="_blank">Harvard Business Review</a>, and was <a href="https://fortune.com/2017/11/16/nvidia-ceo-jensen-huang/" rel="noopener noreferrer" target="_blank">Fortune</a>’s 2017 Businessperson of the Year.He is also an <a href="https://spectrum.ieee.org/ieee-hkn-honor-society-15" target="_self">IEEE–Eta Kappa Nu</a> eminent member.This year’s IEEE Medal of Honor, along with other high-profile IEEE awards, will be presented during the <a href="https://corporate-awards.ieee.org/event/laureate-forum-honors-ceremony-gala/" rel="noopener noreferrer" target="_blank">IEEE Honors Ceremony</a>, to be held in April in New York City. To follow news and updates on IEEE’s most prestigious awards, follow <a href="https://www.linkedin.com/showcase/ieee-awards" rel="noopener noreferrer" target="_blank">IEEE Awards</a> on LinkedIn.

Jan 16, 2026

Video Friday: Bipedal Robot Stops Itself From Falling

<img src="https://spectrum.ieee.org/media-library/prototype-robot-next-to-a-digital-model-both-with-rounded-bodies-and-dome-shaped-heads.png?id=62822757&width=1245&height=700&coordinates=13%2C0%2C13%2C0"/> Video Friday is your weekly selection of awesome robotics videos, collected by your friends at IEEE Spectrum robotics. We also post a weekly calendar of upcoming robotics events for the next few months. Please <a href="mailto:automaton@ieee.org?subject=Robotics%20event%20suggestion%20for%20Video%20Friday">send us your events</a> for inclusion.<h5><a href="https://2026.ieee-icra.org/">ICRA 2026</a>: 1–5 June 2026, VIENNA</h5>Enjoy today’s videos!<div class="horizontal-rule"></div><div style="page-break-after: always"> </div>This is one of the best things I have ever seen. <iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/zklqW-EtVIY?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://publish.illinois.edu/kimlab2020/">Kinetic Intelligent Machine LAB</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="vlvttnvy_mc">After years of aggressive testing and pushing the envelope with U.S. Army and Marine Corps partners, the <a data-linked-post="2650233072" href="https://spectrum.ieee.org/nasa-jpl-team-costar-darpa-subt-urban-circuit-systems-track" target="_blank">Robotic Autonomy in Complex Environments with Resiliency</a> (RACER) program approaches its conclusion. But the impact of RACER will reverberate far beyond the program’s official end date, leaving a legacy of robust autonomous capabilities ready to transform military operations and inspire a new wave of private-sector investment.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/vLVtTNVY_Mc?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.darpa.mil/news/2026/racer-finish-line">DARPA</a> ]<div class="horizontal-rule"></div>Best-looking humanoid yet.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/a_HdbR3g_co?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://kawasakirobotics.com/eu-africa/news/20200714-01/">Kawasaki</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="0hiqs3tbb5g">COSA (Cognitive OS of Agents) is a physical-world-native Agentic OS that unifies high-level cognition with whole-body motion control, enabling humanoid robots to think while acting in real environments. Powered by COSA, Oli becomes the first humanoid agent with both advanced loco-manipulation and high-level autonomous cognition.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/0hIqs3TBb5g?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.limxdynamics.com/en">LimX Dynamics</a> ]Thanks, Jinyan!<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="ls_z60kjvek">The 1X World Model’s latest update is a paradigm shift in robot learning: NEO now uses a physics-grounded video model (World Model) to turn any voice or text prompt into fully autonomous action, even for completely novel tasks and objects NEO has never seen before. By leveraging internet-scale video data fine-tuned on real robot experience, NEO can visualize future actions, predict outcomes, and execute them with humanlike understanding–all without prior examples. This marks the critical first step in NEO being able to collect data on its own to master new tasks all by itself. </blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/lS_z60kjVEk?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.1x.tech/">1X</a> ]<div class="horizontal-rule"></div>I’m impressed by the human who was mocapped for this.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/uuz00OZq_ZA?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://pndbotics.com/">PNDbotics</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="ddnnl4d1kx8">We introduce the GuideData Dataset, a collection of qualitative data, focusing on the interactions between guide dog trainers, visually impaired (BLV) individuals, and their guide dogs. The dataset captures a variety of real-world scenarios, including navigating sidewalks, climbing stairs, crossing streets, and avoiding obstacles. By providing this comprehensive dataset, the project aims to advance research in areas such as assistive technologies, robotics, and human-robot interaction, ultimately improving the mobility and safety of visually impaired people.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/DDNnL4D1kX8?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://guidedogrobot-hgidataset.github.io/">DARoS Lab</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="_v77-uoldiq">Fourier’s desktop Care-Bot prototype is gaining much attention at <a data-linked-post="2674863582" href="https://spectrum.ieee.org/robots-ces-2026" target="_blank">CES 2026</a>! Even though it’s still in the prototype stage, we couldn’t wait to share these adorable and fun interaction features with you.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/_v77-uoLDIQ?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.fftai.com/">Fourier</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="7_a3bfefcje">Volcanic gas measurements are critical for understanding eruptive activity. However, harsh terrain, hazardous conditions, and logistical constraints make near-surface data collection extremely challenging. In this work, we present an autonomous legged robotic system for volcanic gas monitoring, validated through real-world deployments on Mount Etna. The system combines a quadruped robot equipped with a quadrupole mass spectrometer and a modular autonomy stack, enabling long-distance missions in rough volcanic terrain.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/7_a3BFefcJE?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://leggedrobotics.github.io/etna-expedition/">ETH Zurich RSL</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="3fixjy2gwtg">Humanoid and Siemens successfully completed a POC testing humanoid robots in industrial logistics. This is the first step in the broader partnership between the companies. The POC focused on a tote-to-conveyor destacking task within Siemens’s logistics process. HMND 01 autonomously picked, transported, and placed totes in a live production environment during a two-week on-site deployment at the Siemens Electronics Factory in Erlangen.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/3FIXjy2GWTg?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://thehumanoid.ai/">Humanoid</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="qjndiopdnby">Four Growers, a category leader in intelligent ag-tech platforms, developed the GR-200 <a data-linked-post="2650276075" href="https://spectrum.ieee.org/autonomous-robots-plant-tend-and-harvest-entire-crop-of-barley" target="_blank">robotic harvesting</a> platform, powered by FANUC’s LR Mate robot. The system combines AI-driven vision and motion planning to identify and harvest ripe tomatoes with quick precision.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/QJndIoPDnBY?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.fanucamerica.com/case-studies/automating-agriculture-greenhouse-turns-to-robots-for-tomato-harvesting">FANUC</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="nhfu5kha2fw">Columbia Engineers built a robot that, for the first time, is able to learn facial lip motions for tasks such as speech and singing. In a new study published in Science Robotics, the researchers demonstrate how their robot used its abilities to articulate words in a variety of languages, and even sing a song out of its AI-generated debut album, “hello world_.” The robot acquired this ability through observational learning rather than via rules. It first learned how to use its 26 facial motors by watching its own reflection in the mirror before learning to imitate human lip motion by watching hours of YouTube videos.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/nhFU5KHA2fw?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.engineering.columbia.edu/about/news/robot-learns-lip-sync">Columbia</a> ]<div class="horizontal-rule"></div>Roborock has some odd ideas about what lawns are like.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/p0eXOIOZi6Y?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://newsroom.roborock.com/gl/news/ces-2026-roborock-releases-the-world-s-first-robotic-vacuum-with-wheel-leg-architecture-as-it-joins-hands-with-real-madrid-football-club-">Roborock</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="k_sjgiajhes">DEEP Robotics’ quadruped robots demonstrate coordinated multi-module operations under unified command, tackling complex and dynamic firefighting scenarios with agility and precision.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/K_sJGIAjhes?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.deeprobotics.us/">DEEP Robotics</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="njhdpouccoe">Unlike statically stable wheeled platforms, humanoids are dynamically stable, requiring continuous active control to maintain balance and prevent falls. This inherent instability presents a critical challenge for functional safety, particularly in collaborative settings. This presentation will introduce Synapticon’s POSITRON platform, a comprehensive solution engineered to address these safety-critical demands. We will explore how its integrated hardware and software enable robust, certifiable safety functions that meet the highest industrial standards, providing key insights into making the next generation of humanoid robots safe for real-world deployment.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/njHdPOUCcoE?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.synapticon.com/en/products/positron-safety">Synapticon</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="voocufteiaw">The University of California, Berkeley, is world-famous for its AI developments, and one big name behind them is <a data-linked-post="2650253755" href="https://spectrum.ieee.org/ken-goldberg-discusses-telerobots-androids-and-heidegger" target="_blank">Ken Goldberg</a>. Longtime professor and lifelong artist, Ken is all about deep learning while staying true to “good old-fashioned engineering.” Hear Ken talk about his approach to vision and touch for robotic surgeries and how robots will evolve across the board.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/VooCuFTEIaw?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.youtube.com/playlist?list=PLCkt0hth826G9AtnOrQsPbKKD5JmdaMXb">Waymo</a> ]<div class="horizontal-rule"></div>

Jan 15, 2026

How to Gain Footing in AI as the Ground Keeps Shifting

<img src="https://spectrum.ieee.org/media-library/a-young-adult-woman-working-on-her-desktop-computer-in-an-office.jpg?id=62805163&width=1245&height=700&coordinates=0%2C156%2C0%2C157"/> The newly released <a href="https://join.computer.org/ai-developer-career/?Campaign_ID=392" rel="noopener noreferrer" target="_blank">Preparing for a Career as an AI Developer</a> guide from<a href="https://www.computer.org/" rel="noopener noreferrer" target="_blank"> </a>the <a href="https://www.computer.org/" rel="noopener noreferrer" target="_blank">IEEE Computer Society</a> argues that the most durable path to <a href="https://spectrum.ieee.org/tag/artificial-intelligence" target="_self">artificial intelligence</a> jobs is not defined by mastering any single tool or model. Instead, it depends on cultivating a balanced mix of technical fundamentals and human-centered skills—capabilities that machines are unlikely to replace.AI is reshaping the job market faster than most academic programs and employers can keep up with, according to the guide. AI systems now can analyze <a href="https://spectrum.ieee.org/tag/cybercrime" target="_self">cybercrime</a>, predict equipment failures in <a href="https://spectrum.ieee.org/stealthy-startup-aims-to-reinvent-ai-for-manufacturing" target="_self">manufacturing</a>, and generate text, code, and images at scale, leading to <a href="https://spectrum.ieee.org/tech-jobs" target="_self">mass layoffs</a> across much of the technology sector. It has unsettled recent graduates about to enter the job market as well as early-career professionals.Yet the demand for AI expertise remains strong in the <a href="https://spectrum.ieee.org/discover-manager-credit-card-fraud" target="_self">banking</a>, <a href="https://spectrum.ieee.org/robotic-process-automation-gokul-pandy" target="_self">health care</a>, <a href="https://spectrum.ieee.org/abhishek-agrawal-amazon-catalog-ai" target="_self">retail</a>, and pharmaceutical industries, whose businesses are racing to deploy generative AI tools to improve productivity and decision-making—and keep up with the competition.The uneven landscape leaves many observers confused about how best to prepare for a career in a field that is redefining itself. Addressing that uncertainty is the focus of the guide, which was written by <a href="https://www.computer.org/profiles/san-murugesan" rel="noopener noreferrer" target="_blank">San Murugesan</a> and<a href="https://www.wpi.edu/people/faculty/rneamtu" rel="noopener noreferrer" target="_blank"> Rodica Neamtu</a>.Murugesan, an IEEE life senior member, is an adjunct professor at <a href="https://www.westernsydney.edu.au/" rel="noopener noreferrer" target="_blank">Western Sydney University</a>, in Penrith, Australia. Neamtu, an IEEE member, is a professor of teaching and a data-mining researcher at <a href="https://www.wpi.edu/" rel="noopener noreferrer" target="_blank">Worcester Polytechnic Institute</a>, in Massachusetts.The downloadable 24-page PDF outlines what aspiring AI professionals should focus on, which skills are most likely to remain valuable amid rapid automation, and why AI careers are increasingly less about building algorithms in isolation and more about applying them thoughtfully across domains.The guide emphasizes adaptability as the defining requirement for entering the field, rather than fluency in any particular programming language or framework.<h2>Why AI careers are being redefined</h2>AI systems perform tasks that once required human intelligence. What distinguishes the current situation from when AI was introduced, the authors say, is not just improved performance but also expanded scope. Pattern recognition, reasoning, optimization, and machine learning are now used across nearly every sector of the economy.Although automation is expected to reduce the number of human roles in production, office support, customer service, and related fields, demand is rising for people who can design, guide, and integrate AI systems, Murugesan and Neamtu write.The guide cites surveys of executives about AI’s effect on their hiring and retention strategies, including those conducted by <a href="https://www.mckinsey.com/" rel="noopener noreferrer" target="_blank">McKinsey & Co.</a> The reports show staffing shortages in advanced IT and data analytics, as well as applicants’ insufficient critical thinking and creativity: skills that are difficult to automate.The authors frame the mismatch as an opportunity for graduates and early-career professionals to prepare strategically, focusing on capabilities that are likely to remain relevant as AI tools evolve.<h2>Developing complementary skills</h2>The strategic approach aligns with advice from <a href="https://www.csail.mit.edu/person/neil-thompson" rel="noopener noreferrer" target="_blank">Neil Thompson</a>, director of FutureTech research at <a href="https://www.mit.edu/" rel="noopener noreferrer" target="_blank">MIT</a>’s <a href="https://www.csail.mit.edu/" rel="noopener noreferrer" target="_blank">Computer Science and Artificial Intelligence Laboratory</a>, who was quoted in the guide. Thompson encourages workers to develop skills that complement AI rather than compete with it.“When we see rapid technological progress like this, workers should focus on skills and occupations that apply AI to adjacent domains,” he says. “Applying AI in science, in particular, has enormous potential right now and the capacity to unlock significant benefits for humanity.”<h2>The technical foundation still matters</h2>Adaptability, the guide stresses, is not a substitute for technical rigor. A viable AI career still requires a strong foundation in data,<a href="https://spectrum.ieee.org/tag/machine-learning" target="_self"> machine learning</a>, and computing infrastructure.Core knowledge areas include data structures, large-scale data handling, and tools for data manipulation and analysis, the authors say.Foundational machine-learning concepts, such as supervised and unsupervised learning, neural networks, and reinforcement learning, remain essential, they say.Because many AI systems depend on scalable computing, familiarity with cloud platforms such as <a href="https://aws.amazon.com/" rel="noopener noreferrer" target="_blank">Amazon Web Services</a>,<a href="https://cloud.google.com/" rel="noopener noreferrer" target="_blank"> Google Cloud</a>, and <a href="https://azure.microsoft.com/en-us/pricing/purchase-options/azure-account/search?ef_id=_k_Cj0KCQiAxonKBhC1ARIsAIHq_lubfghGwriw2ZaILxIzXBmcfubq9iMM4PqzkwhBGATSfLfbOtvCx7YaAjz5EALw_wcB_k_&OCID=AIDcmmfq865whp_SEM__k_Cj0KCQiAxonKBhC1ARIsAIHq_lubfghGwriw2ZaILxIzXBmcfubq9iMM4PqzkwhBGATSfLfbOtvCx7YaAjz5EALw_wcB_k_&gad_source=1&gad_campaignid=21503043035&gbraid=0AAAAADcJh_sCA_X2u0APALLSysANK0Vhk&gclid=Cj0KCQiAxonKBhC1ARIsAIHq_lubfghGwriw2ZaILxIzXBmcfubq9iMM4PqzkwhBGATSfLfbOtvCx7YaAjz5EALw_wcB" rel="noopener noreferrer" target="_blank">Microsoft Azure</a> is important, according to the guide’s authors.Mathematics underpins all of it. Linear algebra, calculus, and probabilities form the basis of most AI algorithms.<a href="https://spectrum.ieee.org/top-programming-languages-2025" target="_self">Python</a> has emerged as the dominant language for building and experimenting with models.<h2>From algorithms to frameworks</h2>The authors highlight the value of hands-on experience with widely used development frameworks. <a href="https://pytorch.org/" rel="noopener noreferrer" target="_blank">PyTorch</a>, developed by <a href="https://www.meta.ai/" rel="noopener noreferrer" target="_blank">Meta AI</a>, is commonly used for prototyping deep-learning models in academia and industry. <a href="https://scikit-learn.org/stable/" rel="noopener noreferrer" target="_blank">Scikit-learn</a> provides open-source tools for classification, regression, and clustering within the Python ecosystem.“When we see rapid technological progress like this, workers should focus on skills and occupations that apply AI to adjacent domains. —Neil Thompson, MIT<a href="https://www.tensorflow.org/" target="_blank">TensorFlow</a>, a software library for machine learning and AI created by <a href="https://www.google.com/?zx=1767047914659&no_sw_cr=1" rel="noopener noreferrer" target="_blank">Google</a>, supports building and deploying machine-learning systems at multiple levels of abstraction.The authors emphasize that such tools matter less as résumé keywords than as vehicles for understanding how models behave within real-world constraints.<h2>Soft skills as career insurance</h2>Because AI projects often involve ambiguous problems and interdisciplinary teams, soft skills play an increasingly central role, according to the guide. Critical thinking and problem-solving are essential, but communication has become more important, the authors say. Many AI professionals must explain system behavior, limitations, and risks to nontechnical stakeholders.Neamtu describes communication and contextual thinking as timeless skills that grow more valuable as automation expands, particularly when paired with leadership, resilience, and a commitment to continuous learning.Murugesan says technical depth must be matched with the ability to collaborate and adapt.<h2>Experience before titles</h2>The guide recommends that students consider work on research projects in college, as well as paid internships, for exposure to real AI workflows and job roles with hands-on experience.Building an AI project portfolio is critical. Open-source repositories on platforms such as <a href="https://github.com/" rel="noopener noreferrer" target="_blank">GitHub</a> allow newcomers to demonstrate applied skills including work on AI security, bias mitigation, and deepfake detection. The guide recommends staying current by reading academic papers, taking courses, and attending conferences. Doing so can help students get a solid grounding in the basics and remain relevant in a fast-moving field after beginning their career.<h2>Entry-level roles that open doors</h2>Common starting positions include AI research assistant, junior machine-learning engineer, and junior data analyst. The roles typically combine support tasks with opportunities to help develop models, preprocess data, and communicate results through reports and visualizations, according to the guide.Each starting point reinforces the guide’s central message: AI careers are built through collaboration and learning, not merely through isolated technical brilliance.<h2>Curiosity as a long-term strategy</h2>Murugesan urges aspiring AI professionals to embrace continuous learning, seek mentors, and treat mistakes as part of the learning process.“Always be curious,” he says. “Learn from failure. Mistakes and setbacks are part of the journey. Embrace them and persist.”Neamtu echoes that perspective, noting that AI is likely to affect nearly every profession, making passion for one’s work and compatibility with organizational aims more important than chasing the latest technology trend.In a field where today’s tools can become obsolete in a year, the guide’s core argument is simple: The most future-proof AI career is built not on what you know now but on how well you continue learning when things change.

Jan 15, 2026

Lessons for Your Career From 2025

<img src="https://spectrum.ieee.org/media-library/an-illustration-of-stylized-people-wearing-business-casual-clothing.webp?id=62078755&width=1245&height=700&coordinates=0%2C112%2C0%2C113"/> This article is crossposted from IEEE Spectrum’s careers newsletter. <a href="https://engage.ieee.org/Career-Alert-Sign-Up.html" rel="noopener noreferrer" target="_blank">Sign up now</a> to get insider tips, expert advice, and practical strategies, written in partnership with tech career development company <a href="https://jointaro.com/" rel="noopener noreferrer" target="_blank">Taro</a> and delivered to your inbox for free!As we enter 2026, we’re taking a look back at the top pieces of advice we shared in the Career Alert newsletter last year. Whether you’re looking for a new job or seeking strategies to excel in your current role, read on for the three most popular recommendations that could help advance your career. <h2><a href="https://spectrum.ieee.org/getting-past-procastination" target="_blank">1. Getting Past Procrastination</a></h2>Across a decade working at hypergrowth tech companies like Meta and Pinterest, I constantly struggled with procrastination. I’d be assigned an important project, but I simply couldn’t get myself to start it. The source of my distraction varied—I would constantly check my email, read random documentation, or even scroll through my social feeds. But the result was the same: I felt a deep sense of dread that I was not making progress on the things that mattered.At the end of the day, time is the only resource that matters. With every minute, you are making a decision about how to spend your life. Most of the ways people spend their time are ineffective. Especially in the tech world, our tasks and tools are constantly changing, so we must be able to adapt. What separates the best engineers from the rest of the pack is that they create systems that allow them to be consistently productive.Here’s the core idea that changed my perspective on productivity: Action leads to motivation, not the other way around. You should not check your email or scroll Instagram while you wait for motivation to “hit you.” Instead, just start doing something, anything, that makes progress toward your goal, and you’ll find that motivation will follow.…Read the full newsletter <a href="https://spectrum.ieee.org/getting-past-procastination" target="_self">here</a>.<h2><a href="https://spectrum.ieee.org/improving-interview-odds" target="_blank">2. Improve Your Chances of Landing That Job Interview</a></h2>One of my close friends is a hiring manager at Google. She recently posted about an open position on her team and was immediately overwhelmed with applications. We’re talking about thousands of applicants within days.What surprised me most, however, was the horrendous quality of the average submission. Most applicants were obviously unqualified or had concocted entirely fake profiles. The use of generative AI to automatically fill out (and, in some cases, even submit) applications is harmful to everyone; employers are unable to filter through the noise, and legitimate candidates have a harder time getting noticed—much less advancing to an interview.So how can job seekers stand out among the deluge of candidates? When there are hundreds or thousands of applicants, the best way to distinguish yourself is by leveraging your network.With AI, anyone with a computer can trivially apply to thousands of jobs. On the other hand, people are restricted by Dunbar’s number—the idea that humans can maintain stable social relationships with only about 150 people. Being one of those 150 people is harder, but it also carries more weight than a soulless job application.Read the full newsletter <a href="https://spectrum.ieee.org/improving-interview-odds" target="_self">here</a>.<h2><a href="https://spectrum.ieee.org/learning-coding-age-of-ai" target="_blank">3. Learning to Code Still Matters in the Age of AI</a></h2>Cursor, the AI-native code editor, <a href="https://x.com/amanrsanger/status/1916968123535880684" rel="noopener noreferrer" target="_blank">recently reported</a> that it writes nearly a billion lines of code daily. That’s one billion lines of production-grade code accepted by users every single day. If we generously assume that a strong engineer writes a thousand lines of code in a day, Cursor is doing the equivalent work of a million developers. (For context, while working at Pinterest and Meta, I’d typically write less than 100 lines of code per day.)There are only about 25 million software developers worldwide! Naively, it appears that Cursor is making a meaningful percentage of coders obsolete.This begs the question: Is it even worth learning to code anymore?The answer is a resounding “yes.” The above fear-based analysis of Cursor misses several important points.…Read the full newsletter <a href="https://spectrum.ieee.org/learning-coding-age-of-ai" target="_self">here</a>. —Rahul

Jan 14, 2026

The Ultimate 3D Integration Would Cook Future GPUs

<img src="https://spectrum.ieee.org/media-library/close-up-of-an-amd-chip-s-die-face.jpg?id=62725707&width=1245&height=700&coordinates=0%2C187%2C0%2C188"/> Peek inside the package of AMD’s or Nvidia’s <a href="https://spectrum.ieee.org/amd-mi300" target="_blank">most advanced AI products</a>, and you’ll find a familiar arrangement: The GPU is flanked on two sides by <a href="https://spectrum.ieee.org/processing-in-dram-accelerates-ai" target="_self">high-bandwidth memory (HBM)</a>, the most advanced memory chips available. These memory chips are placed as close as possible to the computing chips they serve in order to cut down on the biggest bottleneck in AI computing—the <a href="https://spectrum.ieee.org/trillion-transistor-gpu" target="_blank">energy and delay</a> in getting billions of bits per second from memory into logic. But what if you could bring computing and memory even closer together by stacking the HBM on top of the GPU? Imec recently explored this scenario using advanced thermal simulations, and the answer—delivered in December at the <a href="https://www.ieee-iedm.org/" target="_blank">2025 IEEE International Electron Device Meeting</a> (IEDM)—was a bit grim. 3D stacking doubles the operating temperature inside the GPU, rendering it inoperable. But the team, led by Imec’s <a href="https://www.linkedin.com/in/james-myers-3163a1143/?originalSubdomain=uk" target="_blank">James Myers</a>, didn’t just give up. They identified several engineering optimizations that ultimately could whittle down the temperature difference to nearly zero.<h2>2.5D and 3D Advanced Packaging</h2>Imec started with a thermal simulation of a GPU and four HBM dies as you’d find them today, inside what’s called a 2.5D package. That is, both the GPU and the HBM sit on substrate called an interposer, with minimal distance between them. The two types of chips are linked by thousands of micrometer-scale copper interconnects built into the interposer’s surface. In this configuration, the model GPU consumes 414 watts and reaches a peak temperature of just under 70 °C—typical for a processor. The memory chips consume an additional 40 W or so and get somewhat less hot. The heat is removed from the top of the package by the kind of <a href="https://spectrum.ieee.org/data-center-liquid-cooling" target="_blank">liquid cooling </a>that’s become common in new AI data centers.RELATED: <a href="https://spectrum.ieee.org/hot-chips" target="_blank">Future Chips Will Be Hotter Than Ever</a>“While this approach is currently used, it does not scale well for the future—especially as it blocks two sides of the GPU, limiting future GPU-to-GPU connections inside the package,” <a href="https://www.linkedin.com/in/yukai-chen-88948675/" target="_blank">Yukai Chen</a>, a senior researcher at Imec, told engineers at IEDM. In contrast, “the 3D approach leads to higher bandwidth, lower latency.… The most important improvement is the package footprint.”Unfortunately, as Chen and his colleagues found, the most straightforward version of stacking, simply putting the HBM chips on top of the GPU and adding a block of blank silicon to fill in a gap at the center, shot up temperatures in the GPU to a scorching 140 °C—well past a typical GPU’s 80 °C limit.<h2>System Technology Co-optimization</h2>The Imec team set about trying a number of technology and system optimizations aimed at lowering the temperature. The first thing they tried was throwing out a layer of silicon that was now redundant. To understand why, you have to first get a grip on what HBM really is.This form of memory is a stack of as many as 12 high-density DRAM dies. Each has been thinned down to tens of micrometers and is shot through with vertical connections. These thinned dies are stacked one atop another and connected by tiny<a href="https://spectrum.ieee.org/hybrid-bonding" target="_blank"> balls of solder</a>, and this stack of memory is vertically connected to another piece of silicon, called the base die. The base die is a logic chip designed to multiplex the data—pack it into the limited number of wires that can fit across the millimeter-scale gap to the GPU.But with the HBM now on top of the GPU, there’s no need for such a data pump. Bits can flow directly into the processor without regard for how many wires happen to fit along the side of the chip. Of course, this change means moving the memory control circuits from the base die into the GPU and therefore changing the processor’s floorplan, says Myers. But there should be ample room, he suggests, because the GPU will no longer need the circuits used to demultiplex incoming memory data.RELATED: <a href="https://spectrum.ieee.org/special-reports/keeping-chips-cool/" target="_blank">The Hot, Hot Future of Chips</a>Cutting out this middleman of memory cooled things down by only a little less than 4 °C. But, importantly, it should massively boost the bandwidth between the memory and the processor, which is important for another optimization the team tried—slowing down the GPU.That might seem contrary to the whole purpose of better AI computing, but in this case, it’s an advantage. Large language models are what are called “memory-bound” problems. That is, memory bandwidth is the main limiting factor. But Myers’s team estimated 3D stacking HBM on the GPU would boost bandwidth fourfold. With that added headroom, even slowing the GPU’s clock by 50 percent still leads to a performance win, while cooling everything down by more than 20 °C. In practice, the processor might not need to be slowed down quite that much. Increasing the clock frequency to 70 percent led to a GPU that was only 1.7 °C warmer, Myers says.<h2>Optimized HBM</h2>Another big drop in temperature came from making the HBM stack and the area around it more conductive. That included merging the four stacks into two wider stacks, thereby eliminating a heat-trapping region; thinning out the top—usually thicker—die of the stack; and filling in more of the space around the HBM with blank pieces of silicon to conduct more heat. <img alt="Chart showing 7 steps with temperatures listed on them." class="rm-shortcode" data-rm-shortcode-id="4fb0c31f6fc41c1371c3b675d355c9b9" data-rm-shortcode-name="rebelmouse-image" id="4a274" loading="lazy" src="https://spectrum.ieee.org/media-library/chart-showing-7-steps-with-temperatures-listed-on-them.jpg?id=62861305&width=980"/> Imec explored seven steps to reduce the thermal penalty of stacking memory on GPUs.Source image: ImecWith all of that, the stack now operated at about 88 °C. One final optimization brought things back to near 70 °C. Generally, some 95 percent of a chip’s heat is removed from the top of the package, where in this case water carries the heat away. But adding similar cooling to the underside as well drove the stacked chips down a final 17 °C.Although the research presented at IEDM shows it might be possible, HBM-on-GPU isn’t necessarily the best choice, Myers says. “We are simulating other system configurations to help build confidence that this is or isn’t the best choice,” he says. “GPU-on-HBM is of interest to some in industry,” because it puts the GPU closer to the cooling. But it would likely be a more complex design, because the GPU’s power and data would have to flow vertically through the HBM to reach it.

Jan 14, 2026

Stretchable OLEDs Just Got a Huge Upgrade

<img src="https://spectrum.ieee.org/media-library/a-bi-axially-stretched-exciplex-assisted-phosphorescent-film-deposited-on-a-small-stretchable-substrate.jpg?id=62303035&width=1245&height=700&coordinates=0%2C187%2C0%2C188"/> Wearable displays are catching up with phones and smartwatches. For decades, engineers have sought OLEDs that can bend, twist, and stretch while maintaining bright and stable light. These displays could be integrated into a new class of devices—woven into clothing fabric, for example, to show real-time information, like a runner’s speed or heart rate, without breaking or dimming.But engineers have always encountered a trade-off: The more you stretch these materials, the dimmer they become. Now, a group co-led by <a href="https://drexel.edu/engineering/about/faculty-staff/G/gogotsi-yury/" rel="noopener noreferrer" target="_blank">Yury Gogotsi</a>, a materials scientist at Drexel University in Philadelphia, has found a way around the problem by employing a special class of materials called <a href="https://spectrum.ieee.org/why-mxenese-matter" target="_self">MXenes</a>—which Gogotsi helped discover—that maintain brightness while being significantly stretched.The team developed an OLED that can stretch to twice its original size while keeping a steady glow. It also converts electricity into light more efficiently than any stretchable OLED before it, reaching a record 17 percent external quantum efficiency—a measure of how efficiently a device turns electricity into light.<h2>The “Perfect Replacement”</h2>Gogotsi didn’t have much experience with OLEDs when, about five years ago, he teamed up with <a href="https://www.pnel.snu.ac.kr/professor-intro" rel="noopener noreferrer" target="_blank">Tae-Woo Lee</a>, a materials scientist at Seoul National University, to develop better flexible OLEDs, driven by the ever-increasing use of <a href="https://spectrum.ieee.org/from-foldable-phones-to-stretchy-screens" target="_self">flexible electronics</a> like foldable phones.Traditionally, the displays are built from multiple stacked layers. At the base, a cathode supplies electrons that enter the adjacent organic layers, which are designed to conduct this charge efficiently. As the electrons move through these layers, they meet positive charge injected by an indium tin oxide (ITO) film. The moment these charges combine, the organic material releases energy as light, creating the illuminated pixels that make up the image. The entire structure is sealed with a glass layer on top.The ITO film—adhered to the glass—serves as the anode, allowing current to pass through the organic layers without blocking the generated light. “But it’s brittle. It’s ceramic, basically,” so it works well for flat surfaces, but can’t be bent, Gogotsi explains. There have been attempts to engineer flexible OLEDs many times before, but they failed to meaningfully overcome both flexibility and brightness limitations.Gogotsi’s students started by creating a transparent, conducting film out of a MXene, a type of ultrathin and flexible material with metal-like conductivity. The material is unique in its inherent ability to bend because it’s made from many two-dimensional sheets that can slide relative to each other without breaking. The film—only 10 nanometers thick—“appeared to be this perfect replacement for ITO,” Gogotsi says. Through experimentation, Gogotsi and Lee’s shared team found that a mix of the MXene and silver nanowire would actually stretch the most while maintaining stability. “We were able to double the size, achieving 200 percent stretching without losing performance,” Gogotsi says. <img alt="A bi-axially twisted exciplex-assisted phosphorescent film deposited on a small stretchable substrate." class="rm-shortcode" data-rm-shortcode-id="3e57d12160bbdd4ee5708c032d43b9fe" data-rm-shortcode-name="rebelmouse-image" id="2a744" loading="lazy" src="https://spectrum.ieee.org/media-library/a-bi-axially-twisted-exciplex-assisted-phosphorescent-film-deposited-on-a-small-stretchable-substrate.jpg?id=62303071&width=980"/> The new material can also be twisted without losing its glow.Source image: Huanyu Zhou, Hyun-Wook Kim, et al.And the new MXene film was not only more flexible than ITO but also increased brightness by almost an order of magnitude by making the contact between the topmost light-emitting organic layer and the film more efficient. Unlike ITO, the surface of MXenes can be chemically adjusted to make it easier for electrons to move from the electrode into the light-emitting layer. This more efficient electron flow significantly increases the brightness of the display, as evidenced by an external quantum efficiency of 17 percent, which the group claims is a record for stretchable OLEDs.“Achieving those numbers in intrinsically stretchable OLEDs under substantial stretching is quite significant,” says <a href="https://ee.kaist.ac.kr/en/professor/12225/" target="_blank">Seunghyup Yoo</a>, who runs the Integrated Organic Electronics Laboratory at South Korea’s KAIST. An external quantum efficiency of 20 percent is an important benchmark for this kind of device because it is the upper limit of efficiency dictated by the physical properties of light generation, Yoo explains.To increase illumination, the researchers went beyond working with MXene. Lee’s group developed two additional organic layers to add into the middle of their OLED—one that directs positive charges to the light-emitting layer, ensuring that electricity is used more efficiently, and one that recycles wasted energy that would normally be lost, boosting overall brightness.Together, the MXene layer and two organic layers allow for a notably bright and stable OLED, even when stretched. Gogotsi thinks the subsequent OLED is “very successful” because it combines both brightness and stretchability, while, historically, engineers have only been able to achieve one or the other. “The performance that they are able to achieve in this work is an important advancement,” says <a href="https://pme.uchicago.edu/faculty/sihong-wang" target="_blank">Sihong Wang</a>, a molecular engineer at the University of Chicago who also develops stretchable OLED materials. Wang also notes that the 200 percent stretchability that Gogotsi’s group attained is beyond robust for wearable applications.<h2>Wearables and Health Care</h2>A stretchable OLED that maintains its brightness has uses in many settings, including industrial environments, robotics, wearable clothing and devices, and communications, Gogotsi says, although he’s most excited about its adoption in health-monitoring devices. He sees a near future in which displays for diagnostics and treatment become embedded in clothing or “epidermal electronics,” comparing their function to smartwatches. Before these displays can come to market, however, stability issues inherent to all stretchable OLEDs need to be solved, Wang says. Current materials are not able to sustain light emissions for long enough to serve customers in the ways they require. Finding housings to protect them is also a problem. “You need a stretchable encapsulation material that can protect the central device without allowing oxygen and moisture to permeate,” Wang says.Yoo agrees: He says it’s a tough problem to solve because the best protective layers are rigid and not very stretchable. He notes yet another challenge in the way of commercialization, which is “developing stretchable displays that do not exhibit image distortion.”Regardless, Gogotsi is excited about the future of stretchable OLEDs. “We started with computers occupying the room, then moved to our desktops, then to laptops, then we got smartphones and iPads, but still we carry stuff with us,” he says. “Flexible displays can be on the sleeve of your jacket. They can be <a href="https://spectrum.ieee.org/rollable-smartphone" target="_self">rolled into a tube</a> or folded and put in your pocket. They can be everywhere.”

Jan 13, 2026

Meet the Two Members Petitioning to Be President-Elect Candidates

<img src="https://spectrum.ieee.org/media-library/gerardo-barbosa-and-timothy-lee-smiling.jpg?id=62681379&width=1245&height=700&coordinates=0%2C62%2C0%2C63"/> The IEEE Board of Directors has received petition intentions from IEEE Senior Member Gerardo Barbosa and IEEE Life Senior Member Timothy T. Lee as candidates for 2027 IEEE president-elect. The petitioners are listed in alphabetical order and indicate no preference.The winner of this year’s election will serve as IEEE president in 2028. For more information about the petitioners and Board-nominated candidates, visit <a href="https://ieee.org/pe27" rel="noopener noreferrer" target="_blank">ieee.org/pe27</a>. You can sign their petitions at <a href="https://ieee.org/petition" rel="noopener noreferrer" target="_blank">ieee.org/petition</a>.Signatures for IEEE president-elect candidate petitions are due 10 April at 12:00 p.m. EST/16:00 p.m. UTC. <h2>IEEE Senior Member Gerardo Barbosa</h2> <img alt="Gerardo Barbosa smiling in a suit jacket." class="rm-shortcode" data-rm-shortcode-id="5b9185cf74fac18986516cc526aba4d0" data-rm-shortcode-name="rebelmouse-image" id="afcc0" loading="lazy" src="https://spectrum.ieee.org/media-library/gerardo-barbosa-smiling-in-a-suit-jacket.jpg?id=62681502&width=980"/> Gerardo SosaBarbosa is an expert in information technology management and technology commercialization, with a career spanning innovation, entrepreneurship,and an international perspective. He began his career designing radio-frequency identification systems for real-time asset tracking and inventory management. In 2014 he founded CLOUDCOM, a software company that develops enterprise software to improve businesses’ billing and logistics operations, and serves as its CEO. Barbosa’s IEEE journey began in 2009 at the IEEE Monterrey (Mexico) Section, where he served as chair and treasurer. He led grassroots initiatives with students and young professionals. His leadership positions in IEEE Region 9 include technical activities chair and treasurer. As the 2019—2020 vice chair and 2021—2023 treasurer of IEEE Member and Geographic Activities, Barbosa became recognized as a trusted, data-driven, and collaborative leader. He has been a member of the IEEE Finance Committee since 2021 and is now its chair due to his role as IEEE treasurer on the IEEE Board of Directors. He is deeply committed to the responsible stewardship of IEEE’s global resources, ensuring long-term financial sustainability in service of IEEE’s mission. <h2>IEEE Life Senior Member Timothy T. Lee</h2> <img alt="Timothy Lee smiling." class="rm-shortcode" data-rm-shortcode-id="eb380f867e6ea01f954da7e70e87dfce" data-rm-shortcode-name="rebelmouse-image" id="9bbac" loading="lazy" src="https://spectrum.ieee.org/media-library/timothy-lee-smiling.jpg?id=62681515&width=980"/> Nikon/CESLee is a Technical Fellow at Boeing in Southern California with expertise in microelectronics and advanced 2.5D and 3D chip packaging for AI workloads, 5G, and SATCOM systems for aerospace platforms. He leads R&D projects, including work funded by the Defense Advanced Research Projects Agency. He previously held leadership roles at MACOM Technology Solutions and COMSAT Laboratories. Lee was the 2015 president of the IEEE Microwave Theory and Technology Society. He has served on the IEEE Board of Directors as 2025 IEEE-USA president and 2021–2022 IEEE Region 6 director. He has also been a member of several IEEE committees including Future Directions, Industry Engagement, and New Initiatives. His vision is to deliver societal value through trust, integrity, ownership, innovation, and customer focus, while strengthening the IEEE member experience. Lee also wants to work to prepare members for AI-enabled work in the future.He earned bachelor’s and master’s degrees in electrical engineering from MIT and a master’s degree in systems architecting and engineering from the University of Southern California in Los Angeles.

Jan 13, 2026

This $5,200 Conductive Suit Could Make Power-Line Work Safer

<img src="https://spectrum.ieee.org/media-library/a-man-wearing-a-head-to-toe-suit-posing-for-a-photo-and-displaying-the-strapping-system-that-connects-his-cloves-to-his-jacket.jpg?id=62695027&width=1245&height=700&coordinates=0%2C187%2C0%2C188"/> In 2018, Justin Kropp was working on a transmission circuit in Southern California when disaster struck. Grid operators had earlier shut down the 115-kilovolt circuit, but six high-voltage lines that shared the corridor were still operating, and some of their power snuck onto the deenergized wires he was working on. That rogue current shot to the ground through Kropp’s body and his elevated work platform, killing the 32-year-old father of two.“It went in both of his hands and came out his stomach, where he was leaning against the platform rail,” says Justin’s father, <a href="https://www.linkedin.com/in/barry-kropp-65140676/" target="_blank">Barry Kropp</a>, who is himself a retired line worker. “Justin got hung up on the wire. When they finally got him on the ground, it was too late.” <iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/2SJu2Jp4fSs?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>Budapest-based Electrostatics makes conductive suits that protect line workers from unexpected current. Electrostatics Justin’s accident was caused by induction: a hazard that occurs when an electric or magnetic field causes current to flow through equipment whose intended power supply has been cut off. Safety practices seek to prevent such induction shocks by grounding all conductive objects in a work zone, giving electricity alternative paths. But accidents happen. In Justin’s case, his platform unexpectedly swung into the line before it could be grounded.<h2>Conductive Suits Protect Line Workers</h2>Adding a layer of defense against induction injuries is the motivation behind Budapest-based <a href="https://www.linkedin.com/company/electrostatics/?originalSubdomain=hu" target="_blank">Electrostatics’</a> specialized conductive jumpsuits, which are designed to protect against burns, cardiac fibrillation, and other ills. “If my boy had been wearing one, I know he’d be alive today,” says the elder Kropp, who purchased a line-worker safety training business after Justin’s death. The Mesa, Ariz.–based company, Electrical Safety Consulting International (<a href="https://esci.net/" target="_blank">ESCI</a>), now distributes those suits. <img alt="The lower half of a man\u2019s legs clothed in pants and socks that are connected by straps " class="rm-shortcode" data-rm-shortcode-id="905fe340c107577430765b0fb26ea544" data-rm-shortcode-name="rebelmouse-image" id="17b47" loading="lazy" src="https://spectrum.ieee.org/media-library/the-lower-half-of-a-man-u2019s-legs-clothed-in-pants-and-socks-that-are-connected-by-straps.jpg?id=62695049&width=980"/>Conductive socks that are connected to the trousers complete the protective suit. BME HVL <a href="https://www.linkedin.com/in/eduardo-ramirez-bettoni-2070a345/" target="_blank">Eduardo Ramirez Bettoni</a>, one of the developers of the suits, dug into induction risk after a series of major accidents in the United States in 2017 and 2018, including Justin Kropp’s. At the time, he was principal engineer for transmission and substation standards at Minneapolis-based <a href="https://corporate.my.xcelenergy.com/s/" target="_blank">Xcel Energy</a>. In talking to Xcel line workers and fellow safety engineers, he sensed that the accident cluster might be the tip of an iceberg. And when he and two industry colleagues scoured data from the U.S. Bureau of Labor Statistics, they found 81 induction accidents between 1985 and 2021 and 60 deaths, which they documented in <a href="https://ieeexplore.ieee.org/document/9925039" target="_blank">a 2022 report</a>.“Unfortunately, it is really common. I would say there are hundreds of induction contacts every year in the United States alone,” says Ramirez Bettoni, who is now technical director of R&D for the Houston-based power-distribution equipment firm <a href="https://www.powellind.com/" target="_blank">Powell Industries</a>. He bets that such “contacts”—exposures to dangerous levels of induction—are increasing as grid operators <a href="https://spectrum.ieee.org/grid-congestion-uk" target="_self">boost grid capacity</a> by <a href="https://spectrum.ieee.org/dynamic-line-rating-grid-congestion" target="_self">squeezing additional circuits</a> into transmission corridors.<h3></h3> <div class="rblad-ieee_in_content"></div>Electrostatics’ suits are an enhancement of the standard protective gear that line workers wear when their tasks involve working close to or even touching energized live lines, or “bare-hands” work. Both are interwoven with conductive materials such as stainless steel threads, which form a Faraday cage that shields the wearer against the lines’ electric fields. But the standard suits have limited capacity to shunt current because usually they don’t need to. Like a bird on a wire, bare-hands workers are electrically floating, rather than grounded, so current largely bypasses them via the line itself.<h2>Induction Safety Suit Design</h2>Backed by a US $250,000 investment from Xcel in 2019, Electrostatics adapted its standard suits by adding low-resistance conductive straps that pass current around a worker’s body. “When I’m touching a conductor with one hand and the other hand is grounded, the current will flow through the straps to get out,” says <a href="http://linkedin.com/in/n%25C3%25A9meth-b%25C3%25A1lint-1a366586?originalSubdomain=hu" target="_blank">Bálint Németh</a>, Electrostatics’ CEO and director of the <a href="https://nfl.vet.bme.hu/en/" target="_blank">High Voltage Laboratory</a> at Budapest University of Technology and Economics. <img alt="A man holds one side of his jacket open revealing conductive straps inside. " class="rm-shortcode" data-rm-shortcode-id="f0ad47dc99d78dcc76b59d05cdf16840" data-rm-shortcode-name="rebelmouse-image" id="c7b7c" loading="lazy" src="https://spectrum.ieee.org/media-library/a-man-holds-one-side-of-his-jacket-open-revealing-conductive-straps-inside.jpg?id=62695052&width=980"/>A strapping system links all the elements of the suit—the jacket, trousers, gloves, and socks—and guides current through a controlled path outside the body. BME HVL The company began selling the suits in 2023, and they have since been adopted by over a dozen transmission operators in the United States and Europe, as well as other countries including Canada, Indonesia, and Turkey. They cost about $5,200 in the United States.Electrostatics’ suits had to meet a crucial design threshold: keeping body exposure below the 6-milliampere “let-go” threshold, beyond which electrocuted workers become unable to remove themselves from a circuit. “If you lose control of your muscles, you’re going to hold onto the conductor until you pass out or possibly die,” says Ramirez Bettoni.The gear, which includes the suit, gloves, and socks, protects against 100 amperes for 10 seconds and 50 A for 30 seconds. It also has insulation to protect against heat created by high current and flame retardants to protect against electric arcs.Kropp, Németh, and Ramirez Bettoni are hoping that developing industry standards for induction safety gear, including <a href="https://knowledge.bsigroup.com/products/standard-performance-specification-for-conductive-clothing-for-industry-applications" target="_blank">ones published in October</a>, will broaden their use. Meanwhile, the recently enacted <a href="https://autl.assembly.ca.gov/system/files/2025-03/ab-365-schiavo.pdf" target="_blank">Justin Kropp Safety Act</a> in California, for which the elder Kropp lobbied, mandates automated defibrillators at power-line work sites. This article was updated on 14 January 2026.

Jan 12, 2026

Researchers Beam Power From a Moving Airplane

<img src="https://spectrum.ieee.org/media-library/silhouette-of-an-aircraft-with-widespread-beams-of-light-shining-down-from-its-body.jpg?id=62704978&width=1245&height=700&coordinates=0%2C261%2C0%2C262"/> On a blustery November day, a Cessna turboprop flew over Pennsylvania at 5,000 meters, in crosswinds of up to 70 knots—nearly as fast as the little plane was flying. But the bumpy conditions didn’t thwart its mission: to wirelessly beam power down to receivers on the ground as it flew by.The test flight marked the first time power has been beamed from a moving aircraft. It was conducted by the Ashburn, Va.-based startup <a href="https://www.overviewenergy.com/" target="_blank">Overview Energy</a>, which emerged from stealth mode in December by announcing the feat.But the greater purpose of the flight was to demonstrate the feasibility of a much grander ambition: to beam power from space to Earth. Overview plans to launch satellites into geosynchronous orbit (GEO) to collect unfiltered solar energy where the sun never sets and then beam this abundance back to humanity. The solar energy would be transferred as near-infrared waves and received by existing solar panels on the ground.The far-flung strategy, known as space-based solar power, has become the subject of both <a href="https://spectrum.ieee.org/electromagnetic-waves" target="_blank">daydreaming</a> and <a href="https://spectrum.ieee.org/power-beaming" target="_blank">serious research</a> over the past decade. Caltech’s Space Solar Power Project launched a <a href="https://spectrum.ieee.org/space-based-solar-power" target="_self">demonstration mission</a> in 2023 that transferred power in space using microwaves. And terrestrial power beaming is coming along too. The U.S. Defense Advanced Research Projects Agency (DARPA) in July 2025 set a <a href="https://spectrum.ieee.org/darpa-optical-wireless-power" target="_self">new record for wirelessly transmitting power</a>: 800 watts over 8.6 kilometers for 30 seconds using a laser beam. But until November, no one had actively beamed power from a moving platform to a ground receiver. <h2>Wireless Power Beaming Goes Airborne</h2>Overview’s test transferred only a sprinkling of power, but it did it with the same components and techniques that the company plans to send to space. “Not only is it the first optical power beaming from a moving platform at any substantial range or power,” says Overview CEO <a href="https://www.linkedin.com/in/marc-berte-90559554/" target="_blank">Marc Berte</a>, “but also it’s the first time anyone’s really done a power beaming thing where it’s all of the functional pieces all working together. It’s the same methodology and function that we will take to space and scale up in the long term.”The approach was compelling enough that power-beaming expert <a href="https://www.linkedin.com/in/paul-jaffe-15594a1/" target="_blank">Paul Jaffe</a><a href="https://www.linkedin.com/in/paul-jaffe-15594a1/" target="_blank"></a> left his job as a program manager at DARPA to join the company as head of systems engineering. Prior to DARPA, Jaffe spent three decades with the <a href="https://www.nrl.navy.mil/" target="_blank">U.S. Naval Research Laboratory</a>.“This actually sounds like it could work.” –Paul JaffeIt was hearing Berte explain Overview’s plan at a conference that helped to convince Jaffe to take a chance on the startup. “This actually sounds like it could work,” Jaffe remembers thinking at the time. “It really seems like it gets around a lot of the showstoppers for a lot of the other concepts. I remember coming home and telling my wife that I almost felt like the problem had been solved. So I thought: Should [I] do something which is almost unheard of—to leave in the middle of being a DARPA program manager—to try to do something else?”For Jaffe, the most compelling reason was in Overview’s solution for space-based solar’s power-density problem. A beam with low power density is safer because it’s not blasting too much concentrated energy onto a single spot on the Earth’s surface, but it’s less efficient for the task of delivering usable solar energy. A higher-density beam does the job better, but then the researchers must engineer some way to maintain safety. <iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/fAnzrDIEOuA?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe> Startup Overview Energy demonstrates how space-based solar power could be beamed to Earth from satellites. Overview Energy<h2>Space-Based Solar Power Makes Waves</h2>Many researchers have settled on microwaves as their beam of choice for wireless power. But, in addition to the safety concerns about shooting such intense waves at the Earth, Jaffe says there’s another problem: Microwaves are part of what he calls the “beachfront property” of the electromagnetic spectrum—a range from 2 to 20 gigahertz that is set aside for many other applications, such as 5G cellular networks. “The fact is,” Jaffe says, “if you somehow magically had a fully operational solar power satellite that used microwave power transmission in orbit today—and a multi-kilometer-scale microwave power satellite receiver on the ground magically in place today—you could not turn it on because the spectrum is not allocated to do this kind of transmission.”Instead, Overview plans to use less-dense, wide-field infrared waves. Existing utility-scale solar farms would be able to receive the beamed energy just like they receive the sun’s energy during daylight hours. So “your receivers are already built,” Berte says. The next major step is a prototype demonstrator for low Earth orbit, after which he hopes to have GEO satellites beaming megawatts of power by 2030 and gigawatts by later that decade.<a href="https://spectrum.ieee.org/space-based-solar-power-2667878868" target="_self">Plenty of doubts</a> about the feasibility of space-based power abound. It is an exotic technology with much left to prove, including the ability to survive orbital debris and the exorbitant cost of launching the power stations. (Overview’s satellite will be built on Earth in a folded configuration, and it will unfold after it’s brought to orbit, according to the company.)“Getting down the cost per unit mass for launch is a big deal,” Jaffe says. “Then, it just becomes a question of increasing the specific power. A lot of the technologies we’re working on at Overview are squarely focused on that.”

Jan 11, 2026

Chilean Telescope Array Gets 145 New Powerful Amplifiers

<img src="https://spectrum.ieee.org/media-library/several-atacama-large-millimeter-submillimeter-array-antennas-pointed-up-at-a-starry-night-sky.jpg?id=62687462&width=1245&height=700&coordinates=0%2C62%2C0%2C63"/> For decades, scientists have observed the cosmos with <a data-linked-post="2650279068" href="https://spectrum.ieee.org/new-antenna-uses-saltwater-to-achieve-multiple-beamsteering-states" target="_blank">radio antennas</a> to visualize the dark, distant regions of the universe. This includes the gas and dust of the <a href="https://www.cfa.harvard.edu/research/topic/interstellar-medium-and-molecular-clouds" target="_blank">interstellar medium</a>, planet-forming disks, and objects that cannot be observed in visible light. In this field, the <a data-linked-post="2650269106" href="https://spectrum.ieee.org/alma-makes-a-highprofile-debut" target="_blank">Atacama Large Millimeter/Submillimeter Array</a> (ALMA) in Chile stands out as one of the world’s most powerful radio telescopes. Using its 66 parabolic antennas, <a href="https://www.almaobservatory.org/en/home/" target="_blank">ALMA</a> observes the millimeter and submillimeter radiation emitted by cold molecular clouds from which new stars are born. <img alt='Universe Today logo; text reads "This post originally appeared on Universe Today."' class="rm-shortcode" data-rm-shortcode-id="0cc58af6fee158c1a01c55a221e47cd0" data-rm-shortcode-name="rebelmouse-image" id="b9dd2" loading="lazy" src="https://spectrum.ieee.org/media-library/universe-today-logo-text-reads-this-post-originally-appeared-on-universe-today.png?id=60568425&width=980"/> Each antenna is equipped with high-frequency receivers for 10 wavelength ranges, 35 to 50 gigahertz and 787 to 950 GHz, collectively known as <a href="https://www.eso.org/public/teles-instr/alma/receiver-bands/" target="_blank">Band 1</a>. Thanks to the <a href="https://www.iaf.fraunhofer.de/en.html" target="_blank">Fraunhofer Institute for Applied Solid State Physics</a> (IAF) and the <a href="https://www.mpifr-bonn.mpg.de/2169/en" target="_blank">Max Planck Institute for Radio Astronomy</a>, ALMA has received an upgrade with the addition of 145 new low-noise amplifiers (LNAs). These amplifiers are part of the facilities’ <a href="https://www.eso.org/public/teles-instr/alma/receiver-bands/" target="_blank">Band 2</a> coverage, ranging from 67 to 116 GHz on the electromagnetic spectrum. This additional coverage will allow researchers to study and gain a better understanding of the universe.In particular, they hope to gain new insights into the “cold interstellar medium”: The dust, gas, radiation, and magnetic fields from which stars are born. In addition, scientists will be able to study planet-forming disks in better detail. Last, but certainly not least, they will be able to study complex organic molecules in nearby galaxies, which are considered precursors to the building blocks of life. In short, these studies will allow astronomers and cosmologists to witness how stars and planetary systems form and evolve, and how the presence of organic molecules can lead to the emergence of life.<h2>Advanced Amplifiers Enhance ALMA Sensitivity</h2>Each LNA includes a series of monolithic microwave integrated circuits (MMICs) developed by Fraunhofer IAF using the semiconducting material indium gallium arsenide. MMICs are based on metamorphic high-electron-mobility transistor technology, a method for creating advanced transistors that are flexible and allow for optimized performance in high-frequency receivers. The addition of LNAs equipped with these circuits will amplify low-noise signals and minimize background noise, dramatically increasing the sensitivity of ALMA’s receivers.<a href="https://www.linkedin.com/in/fabian-thome/" target="_blank">Fabian Thome</a>, head of the subproject at Fraunhofer IAF, explained in an <a href="https://www.iaf.fraunhofer.de/en/media-library/press-releases/145-low-noise-amplifiers-alma-telescopes-band-2.html" target="_blank">IAF press release</a>:<blockquote>“The performance of receivers depends largely on the performance of the first high-frequency amplifiers installed in them. Our technology is characterized by an average noise temperature of 22 K, which is unmatched worldwide.” With the new LNAs, signals can be amplified more than 300-fold in the first step. “This enables the ALMA receivers to measure millimeter and submillimeter radiation from the depths of the universe much more precisely and obtain better data. We are incredibly proud that our LNA technology is helping us to better understand the origins of stars and entire galaxies.”</blockquote>Both Fraunhofer IAF and the Max Planck Institute for Radio Astronomy were commissioned by the <a href="https://www.eso.org/public/" target="_blank">European Southern Observatory</a> to provide the amplifiers. While Fraunhofer IAF was responsible for designing, manufacturing, and testing the MMICs at room temperature, Max Planck was tasked with assembling and qualifying the LNA modules, then testing them in cryogenic conditions. “This is a wonderful recognition of our fantastic collaboration with Fraunhofer IAF, which shows that our amplifiers are not only ‘made in Germany’ but also the best in the world,” said <a href="https://www.mpg.de/397578/radio-astronomy-kramer" target="_blank">Michael Kramer</a>, executive director at the Max Planck Institute for Radio Astronomy.

Jan 10, 2026

Nvidia’s New Rubin Architecture Thrives on Networking

<img src="https://spectrum.ieee.org/media-library/nvidia-s-vera-rubin-ai-supercomputer-made-up-of-several-racks-of-cpus-and-gpus.jpg?id=62697156&width=1245&height=700&coordinates=0%2C62%2C0%2C63"/> Earlier this week, Nvidia surprise-<a href="https://nvidianews.nvidia.com/news/rubin-platform-ai-supercomputer" rel="noopener noreferrer" target="_blank">announced</a> their new Vera Rubin architecture (no relation to the recently unveiled <a href="https://spectrum.ieee.org/vera-rubin-observatory-first-images" target="_self">telescope</a>) at the <a href="https://www.ces.tech/" rel="noopener noreferrer" target="_blank">Consumer Electronics Show</a> in Las Vegas. The new platform, set to reach customers later this year, is advertised to offer a tenfold reduction in inference costs and a fourfold reduction in how many GPUs it would take to train certain models, as compared to Nvidia’s <a href="https://spectrum.ieee.org/nvidia-blackwell" target="_self">Blackwell</a> architecture. The usual suspect for improved performance is the GPU. Indeed, the new Rubin GPU boasts 50 quadrillion floating-point operations per second (petaFLOPS) of 4-bit computation, as compared to 10 petaflops on Blackwell, at least for transformer-based inference workloads like large language models.However, focusing on just the GPU misses the bigger picture. There are a total of six new chips in the Vera Rubin-based computers: the Vera CPU, the Rubin GPU, and four distinct networking chips. To achieve performance advantages, the components have to work in concert, says <a href="https://blogs.nvidia.com/blog/author/giladshainer/" rel="noopener noreferrer" target="_blank">Gilad Shainer</a>, senior vice president of networking at Nvidia.“The same unit connected in a different way will deliver a completely different level of performance,” Shainer says. “That’s why we call it extreme co-design.”<h2>Expanded “in-network compute”</h2>AI workloads, both training and inference, run on large numbers of GPUs simultaneously. “Two years back, inferencing was mainly run on a single GPU, a single box, a single server,” Shainer says. “Right now, inferencing is becoming distributed, and it’s not just in a rack. It’s going to go across racks.”To accommodate these hugely distributed tasks, as many GPUs as possible need to effectively work as one. This is the aim of the so-called <a href="https://spectrum.ieee.org/rf-over-fiber" target="_blank">scale-up network</a>: the connection of GPUs within a single rack. Nvidia handles this connection with their NVLink networking chip. The new line includes the NVLink6 switch, with double the bandwidth of the <a href="https://www.nvidia.com/en-us/data-center/nvlink/" target="_blank">previous version</a><a href="https://www.nvidia.com/en-us/data-center/nvlink/" rel="noopener noreferrer" target="_blank"></a> (3,600 gigabytes per second for GPU-to-GPU connections, as compared to 1,800 GB/s for the NVLink5 switch).In addition to the bandwidth doubling, the scale-up chips also include double the number of SerDes—serializer/deserializers (which allow data to be sent across fewer wires)—and an expanded number of calculations that can be done within the network.“The scale-up network is not really the network itself,” Shainer says. “It’s computing infrastructure, and some of the computing operations are done on the network…on the switch.”The rationale for offloading some operations from the GPUs to the network is twofold. First, it allows some tasks to only be done once, rather than having every GPU have to perform them. A common example of this is the all-reduce operation in AI training. During training, each GPU computes a mathematical operation, called a gradient, on its own batch of data. In order to train the model correctly, all the GPUs need to know the average gradient computed across all batches. Rather than each GPU sending its gradient to every other GPU, and every one of them computing the average, it saves computational time and power for that operation to only happen once, within the network.A second rationale is to hide the time it takes to shuttle data in between GPUs by doing computations on them en route. Shainer explains this via an analogy of a pizza parlor trying to speed up the time it takes to deliver an order. “What can you do if you had more ovens or more workers? It doesn’t help you; you can make more pizzas, but the time for a single pizza is going to stay the same. Alternatively, if you would take the oven and put it in a car, so I’m going to bake the pizza while traveling to you, this is where I save time. This is what we do.”In-network computing is not new to this iteration of Nvidia’s architecture. In fact, it has been in common use since around 2016. But, this iteration adds a broader swath of computations that can be done within the network to accommodate different workloads and different numerical formats, Shainer says.<h2>Scaling out and across</h2>The rest of the networking chips included in the Rubin architecture comprise the so-called scale-out network. This is the part that connects different racks to each other within the data center.Those <a href="https://developer.nvidia.com/blog/inside-the-nvidia-rubin-platform-six-new-chips-one-ai-supercomputer/" rel="noopener noreferrer" target="_blank">chips</a> are the ConnectX-9, a networking interface card; the BlueField-4 a so-called data processing unit, which is paired with two Vera CPUs and a ConnectX-9 card for offloading networking, storage, and security tasks; and finally the Spectrum-6 Ethernet switch, which uses <a href="https://spectrum.ieee.org/co-packaged-optics" target="_self">co-packaged optics</a> to send data between racks. The Ethernet switch also doubles the bandwidth of the previous generations while minimizing jitter—the variation in arrival times of information packets.“Scale-out infrastructure needs to make sure that those GPUs can communicate well in order to run a distributed computing workload, and that means I need a network that has no jitter in it,” he says. The presence of jitter implies that if different racks are doing different parts of the calculation, the answer from each will arrive at different times. One rack will always be slower than the rest, and the rest of the racks, full of costly equipment, sit idle while waiting for that last packet. “Jitter means losing money,” Shainer says.None of Nvidia’s host of new chips are specifically dedicated to connect between data centers, termed “scale-across.” But Shainer argues this is the next frontier. “It doesn’t stop here, because we are seeing the demands to increase the number of GPUs in a data center,” he says. “100,000 GPUs is not enough anymore for some workloads, and now we need to connect multiple data centers together.”

Jan 9, 2026

Sena Kizildemir Simulates Disasters to Prevent Building Collapses

<img src="https://spectrum.ieee.org/media-library/portrait-of-sena-kizildemir-against-a-texture-background-of-cracked-concrete.jpg?id=62694914&width=1245&height=700&coordinates=0%2C62%2C0%2C63"/> When two airplanes <a href="https://spectrum.ieee.org/the-events-of-11-september" target="_blank">hit the World Trade Center</a> in New York City on 11 September 2001, no one could predict how the Twin Towers would react structurally. The commercial jet airliners severed columns and started fires, weakening steel beams, and causing a “pancaking,” progressive collapse.Skyscrapers had not been designed or constructed with that kind of catastrophic structural failure in mind. IEEE Senior Member <a href="https://www.linkedin.com/in/senakizildemir/" rel="noopener noreferrer" target="_blank">Sena Kizildemir</a> is changing that through disaster simulation, one scenario at a time.<h3>Sena Kizildemir</h3> EmployerThornton Tomasetti, in New York CityJob titleProject engineerMember grade Senior memberAlma maters Işik University in Şile and Lehigh University, in Bethlehem, Pa.A project engineer at <a href="https://www.thorntontomasetti.com/resource/applied-science-practice" rel="noopener noreferrer" target="_blank">Thornton Tomasetti</a>’s applied science division in New York, Kizildemir uses simulations to study how buildings fail under extreme events such as impacts and explosions. The simulation results can help designers develop mitigation strategies.“Simulations help us understand what could happen before it occurs in real life,” she says, “to be able to better plan for it.”She loves that her work mixes creativity with solving real-world problems, she says: “You’re creating something to help people. My favorite question to answer is, ‘Can you make this better or easier?’”For her work, the nonprofit <a href="https://www.pwc-ny.org/about-us/" rel="noopener noreferrer" target="_blank">Professional Women in Construction</a> named her one of its <a href="https://members.pwc-ny.org/events/Details/holiday-party-2025-1496059?sourceTypeId=Hub" rel="noopener noreferrer" target="_blank">20 Under 40: Women in Construction</a> for 2025.Kizildemir is passionate about <a data-linked-post="2661603745" href="https://spectrum.ieee.org/ieee-collabratec-mentoring-program" target="_blank">mentoring young engineers</a> and being an IEEE volunteer. She says she has made it her mission to “pack as much impact into my years as possible.”<h2>A bright student in Türkiye</h2>She was born in Istanbul to a father who is a professional drummer and a mother who worked in magazine advertising and sales. Kizildemir and her older brother pursued engineering careers despite neither parent being involved in the field. While she became an expert in civil and mechanical engineering, her brother is an industrial engineer.As a child, she was full of curiosity, she says, interested in figuring out how things were built and how they worked. She loved building objects out of <a href="https://www.lego.com/en-us" rel="noopener noreferrer" target="_blank">Legos</a>, she says, and one of her earliest memories is using them to make miniature houses for ants.After acing an entrance exam, she won a spot in a STEM-focused high school, where she studied mathematics and physics.“Engineering is one of the few careers where you can make a lasting impact on the world, and I plan on mine being meaningful.”During her final year at the high school, she took the nationwide <a href="https://en.wikipedia.org/wiki/Student_Selection_and_Placement_System" target="_blank">YKS</a> (Higher Education Foundations Examination). The test determines which universities and programs—such as medicine, engineering, or law—students can pursue.She received a full scholarship to attend <a href="https://www.isikun.edu.tr/en" target="_blank">Işik University</a> in Şile. Figuring she would study engineering abroad one day, she chose an English-taught program. She says she found that civil engineering best aligned with making the biggest impact on her community and the world.Several of her professors were alumni of <a href="https://www2.lehigh.edu/" target="_blank">Lehigh University</a>, in Bethlehem, Pa., and spoke highly of the school. After earning her bachelor’s degree in civil engineering in 2016, she decided to attend Lehigh, where she earned a full scholarship to its master’s program in civil engineering.<h2>Moving abroad and working the rails</h2>Her master’s thesis focused on investigating root causes of crack propagation, which threatens railroad safety.Repeated wheel-rail loading causes microcracks, leading to metal fatigue, and residual stress results from specialized heating and cooling treatments during the manufacturing of steel rails. Cracks can develop beneath the rail’s surface. Because they’re invisible to the naked eye, such fractures are challenging to detect, Kizildemir says.The project was done in collaboration with the U.S. <a href="https://railroads.dot.gov/" rel="noopener noreferrer" target="_blank">Federal Railroad Administration</a>—part of the <a href="https://www.transportation.gov/" rel="noopener noreferrer" target="_blank">Department of Transportation</a>—which is looking to adjust technical standards and employ mitigation strategies.Kizildemir and five colleagues designed and implemented testing protocols and physics-based simulations to detect cracks earlier and prevent their spread. Their research has given the Railroad Administration insights into structural defects that are being used to revise rail-building guidelines and inspection protocols. The administration published the <a href="https://rosap.ntl.bts.gov/view/dot/76534" rel="noopener noreferrer" target="_blank">first phase</a> of the research in 2024.After graduating in 2018, Kizildemir began a summer internship as a civil engineer at Thornton Tomasetti. She conducted computational modeling using <a href="https://en.wikipedia.org/wiki/Abaqus" rel="noopener noreferrer" target="_blank">Abaqus</a> software for rails subjected to repeated plastic deformation—material that permanently changes shape when under excessive stress—and presented her recommendations for improvement to the company’s management.During her internship, she worked with professors in different fields, including materials behavior and mechanical engineering. The experience, she says, inspired her to pursue a Ph.D. in mechanical engineering at Lehigh, continuing her research with the Railroad Administration. She earned her degree in 2023.She loved the work and the team at Thornton Tomasetti so much, she says, that she applied to work at the company, where she is now a project engineer.<h2>From simulations to real-world applications</h2>Her work focuses on developing finite element models for critical infrastructure and extreme events.Finite modeling breaks complex systems or topics into small elements connected together to numerically simulate real-world situations. She creates computational models of structures enduring realistic catastrophic events, such as a vehicle crashing into a building.She uses simulations to understand how buildings react to attacks such as the one on 9/11, which, she says, is often used as an example of why such research is essential.When starting a project, she and her team review building standards and try to identify new issues not yet covered by them. The team then adapts existing codes and standards, usually developed for well-understood hazards such as earthquakes, wind, and floods, to define simulation parameters.When a new structure is being built, for example, it is not designed to withstand a truck crashing into it. But Kizildemir and her team want to know how the building would react should that happen. They simulate the environments and situations, and they make recommendations based on the results to reduce or eliminate risks of structural failure.Mitigation suggestions include specific strategies to be implemented during project design and construction.Simulations can be created for any infrastructure, Kizildemir says.“I love problems that force me to think differently,” she says. “I want to keep growing.”She says she plans to live by Thornton Tomasetti’s internal motto: “When others say no, we say ‘Here’s how.’”<h2>Joining IEEE and getting more involved</h2>When Kizildemir first heard of IEEE, she assumed it was only for electrical engineers. But after learning how diverse and inclusive the organization is, she joined in 2024. She has since been elevated to a senior member and has become a volunteer. She joined the <a href="https://www.ieee.org/membership-catalog/productdetail/showProductDetailPage.html?product=MEMTEM014" rel="noopener noreferrer" target="_blank">IEEE Technology and Engineering Management Society</a>.She chaired the conference tracks and IEEE-sponsored sessions at the 2024 <a href="https://event.asme.org/Joint-Rail-Conference" rel="noopener noreferrer" target="_blank">Joint Rail Conference</a>, held in Columbia, S.C. She actively contributes to IEEE’s <a href="https://ieee-collabratec.ieee.org/" rel="noopener noreferrer" target="_blank">Collabratec platform</a> and has participated in panel review meetings for senior member elevation applications.She’s also a member of <a href="https://www.asme.org/" rel="noopener noreferrer" target="_blank">ASME</a> and has been volunteering for it since 2023.“Community is what helped get me to where I am today, and I want to pay it forward and make the field better,” she says. “Helping others improves ourselves.”<h2>Looking ahead and giving back</h2>Kizildemir mentors junior engineers at Thornton Thomasetti and is looking to expand her reach through <a href="https://www.ieee.org/membership/mentoring" rel="noopener noreferrer" target="_blank">IEEE’s mentorship programs</a>.“Engineering doesn’t have a gender requirement,” she says she tells girls. “If you’re curious and like understanding how things work and get excited to solve difficult problems, engineering is for you.“Civil engineers don’t just build bridges,” she adds. “There are countless niche areas to be explored. Engineering is one of the few careers where you can make a lasting impact on the world, and I plan on mine being meaningful.”Kizildemir says she wants every engineer to be able to improve their community. Her main piece of advice for recent engineering graduates is that “curiosity, discipline, and the willingness to understand things deeply, to see how things can be done better,” are the keys to success.

Jan 9, 2026

Video Friday: Robots Are Everywhere at CES 2026

<img src="https://spectrum.ieee.org/media-library/a-humanoid-robot-stands-in-a-workshop-surrounded-by-yellow-safety-barriers.png?id=62698082&width=1245&height=700&coordinates=0%2C0%2C0%2C0"/> Video Friday is your weekly selection of awesome robotics videos, collected by your friends at IEEE Spectrum robotics. We also post a weekly calendar of upcoming robotics events for the next few months. Please <a href="mailto:automaton@ieee.org?subject=Robotics%20event%20suggestion%20for%20Video%20Friday">send us your events</a> for inclusion.<h5><a href="https://2026.ieee-icra.org/">ICRA 2026</a>: 1–5 June 2026, VIENNA</h5>Enjoy today’s videos!<div class="horizontal-rule"></div><div style="page-break-after: always"> </div><blockquote class="rm-anchors" id="sd8ivhpji6g">We’re excited to announce the product version of our Atlas® robot. This enterprise-grade humanoid robot offers impressive strength and range of motion, precise manipulation, and intelligent adaptability—designed to power the new industrial revolution.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/sd8ivhpjI6g?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe> <iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/rrUHZKlrxms?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://bostondynamics.com/products/atlas/">Boston Dynamics</a> ]<div class="horizontal-rule"></div>I appreciate the creativity and technical innovation here, but realistically, if you’ve got more than one floor in your house? Just get a second robot. That single-step sunken living room though....<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/SuyBti4YC-Y?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://us.roborock.com/pages/ces-2026">Roborock</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="btjxyrtr9m8">Wow, SwitchBot’s CES 2026 video shows almost as many robots in their fantasy home as I have in my real home.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/btJxyrtR9M8?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.switch-bot.com/pages/events-ces-2026">SwitchBot</a> ]<div class="horizontal-rule"></div>What is happening in robotics right now that I can derive more satisfaction from watching <a data-linked-post="2650252967" href="https://spectrum.ieee.org/building-better-solar-cells-at-robot-speed" target="_blank">robotic process automation</a> than I can from watching yet another <a data-linked-post="2673979476" href="https://spectrum.ieee.org/humanoid-robot-olympics" target="_blank">humanoid</a> video?<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/BkUuepmJdc8?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://new.abb.com/news/detail/132408/cstmr-xiang-piao-piao-increases-production-efficiency-by-40-percent-thanks-to-abb-automated-system">ABB</a> ]<div class="horizontal-rule"></div>Yes, this is definitely a robot I want in close proximity to my life.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/JZllfrHRc4g?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.unitree.com/H2">Unitree</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="xxvtjtorxl0">The video below demonstrates a MenteeBot learning, through mentoring, how to replace a battery in another MenteeBot. No teleoperation is used.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/_bjq90duOcM?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.menteebot.com/">Mentee Robotics</a> ]<div class="horizontal-rule"></div>Personally, I think we should encourage humanoid robots to fall much more often, just so we can see whether they can get up again.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/ywWwcecwiZA?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.agilityrobotics.com/">Agility Robotics</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="mx0dzgtptbg">Achieving long-horizon, reliable clothing manipulation in the real world remains one of the most challenging problems in robotics. This live test demonstrates a strong step forward in embodied intelligence, vision-language-action systems, and real-world robotic autonomy.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/MX0DzGtPtBg?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://mmlab.hk/research/kai0">HKU MMLab</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="626qsq1czoc">Millions of people around the world need assistance with feeding. Robotic feeding systems offer the potential to enhance autonomy and quality of life for individuals with impairments and reduce caregiver workload. However, their widespread adoption has been limited by technical challenges such as estimating bite timing, the appropriate moment for the robot to transfer food to a user’s mouth. In this work, we introduce WAFFLE: Wearable Approach For Feeding with LEarned Bite Timing, a system that accurately predicts bite timing by leveraging wearable sensor data to be highly reactive to natural user cues such as head movements, chewing, and talking.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/626qsQ1CZOc?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://sites.google.com/view/bitetiming/">CMU RCHI</a> ]<div class="horizontal-rule"></div>Humanoid robots are now available as platforms, which is a great way of sidestepping the whole practicality question.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/JyPiC0aArPg?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://pndbotics.com/humanoid">PNDbotics</a> ]<div class="horizontal-rule"></div><blockquote class="rm-anchors" id="ig41e0yqijg">We’re introducing Spatially Enhanced Recurrent Units (SRUs)—a simple yet powerful modification that enables robots to build implicit spatial memories for navigation. Published in the International Journal of Robotics Research (IJRR), this work demonstrates up to +105 percent improvement over baseline approaches, with robots successfully navigating 70+ meters in the real world using only a single forward-facing camera.</blockquote><iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/iG41e0yQIjg?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://michaelfyang.github.io/sru-project-website/">ETHZ RSL</a> ]<div class="horizontal-rule"></div>Looking forward to the <a data-linked-post="2674674488" href="https://spectrum.ieee.org/darpa-triage-challenge-robots" target="_blank">DARPA Triage Challenge</a> this fall!<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/GXNPY0-zwcU?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://www.darpa.mil/research/challenges/darpa-triage-challenge">DARPA</a> ]<div class="horizontal-rule"></div>Here are a couple of good interviews from the Humanoids Summit 2025.<iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/U7wnLqt5FS8?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe> <iframe frameborder="0" height="auto" lazy-loadable="true" scrolling="no" src="https://www.youtube.com/embed/vdCWKUfoaFk?rel=0" style="position:absolute;top:0;left:0;width:100%;height:100%;" width="100%"></iframe>[ <a href="https://humanoidssummit.com/">Humanoids Summit</a> ]<div class="horizontal-rule"></div>

Jan 8, 2026

How AI Accelerates PMUT Design for Biomedical Ultrasonic Applications

Jan 8, 2026

AI Coding Assistants Are Getting Worse

<img src="https://spectrum.ieee.org/media-library/an-ouroboros-against-an-abstract-wavy-background.jpg?id=62650375&width=1245&height=700&coordinates=0%2C187%2C0%2C188"/> In recent months, I’ve noticed a troubling trend with AI <a data-linked-post="2671645555" href="https://spectrum.ieee.org/vibe-coding" target="_blank">coding assistants</a>. After two years of steady improvements, over the course of 2025, most of the core models reached a quality plateau, and more recently, seem to be in decline. A task that might have taken five hours assisted by AI, and perhaps 10 hours without it, is now more commonly taking seven or eight hours, or even longer. It’s reached the point where I am sometimes going back and using older versions of <a data-linked-post="2672500550" href="https://spectrum.ieee.org/large-language-model-performance" target="_blank">large language models</a> (LLMs).I use LLM-generated code extensively in my role as CEO of <a href="https://www.carringtonlabs.com/" target="_blank">Carrington Labs</a>, a provider of predictive-analytics risk models for lenders. My team has a sandbox where we create, deploy, and run AI-generated code without a human in the loop. We use them to extract useful features for model construction, a natural-selection approach to feature development. This gives me a unique vantage point from which to evaluate coding assistants’ performance.<h2>Newer models fail in insidious ways</h2>Until recently, the most common problem with AI coding assistants was poor syntax, followed closely by flawed logic. AI-created code would often fail with a syntax error or snarl itself up in faulty structure. This could be frustrating: The solution usually involved manually reviewing the code in detail and finding the mistake. But it was ultimately tractable.However, recently released LLMs, such as <a data-linked-post="2673933939" href="https://spectrum.ieee.org/gpt-5-trough-of-disillusionment" target="_blank">GPT-5</a>, have a much more insidious method of failure. They often generate code that fails to perform as intended, but which on the surface seems to run successfully, avoiding syntax errors or obvious crashes. It does this by removing safety checks, or by creating fake output that matches the desired format, or through a variety of other techniques to avoid crashing during execution.As any developer will tell you, this kind of silent failure is far, far worse than a crash. Flawed outputs will often lurk undetected in code until they surface much later. This creates confusion and is far more difficult to catch and fix. This sort of behavior is so unhelpful that modern programming languages are deliberately designed to fail quickly and noisily.<h2>A simple test case</h2>I’ve noticed this problem anecdotally over the past several months, but recently, I ran a simple yet systematic test to determine whether it was truly getting worse. I wrote some Python code, which loaded a dataframe, and then looked for a nonexistent column.<blockquote>df = pd.read_csv(‘data.csv’) df['new_column'] = df['index_value'] + 1 #there is no column ‘index_value’</blockquote>Obviously, this code would never run successfully. Python generates an easy-to-understand error message that explains that the column ‘index_value’ cannot be found. Any human seeing this message would inspect the dataframe and notice that the column was missing.I sent this error message to nine different versions of ChatGPT, primarily variations on GPT-4 and the more recent GPT-5. I asked each of them to fix the error, specifying that I wanted completed code only, without commentary.This is, of course, an impossible task—the problem is the missing data, not the code. So the best answer would be either an outright refusal, or failing that, code that would help me debug the problem. I ran 10 trials for each model and classified the output as helpful (when it suggested the column is probably missing from the dataframe), useless (something like just restating my question), or counterproductive (for example, creating fake data to avoid an error).GPT-4 gave a useful answer every one of the 10 times that I ran it. In three cases, it ignored my instructions to return only code, explained that the column was likely missing from my dataset, and that I would have to address it there. In six cases, it tried to execute the code but added an exception that would either throw up an error or fill the new column with an error message if the column couldn’t be found (the 10th time, it simply restated my original code).<blockquote>This code will add 1 to the ‘index_value’ column from the dataframe ‘df’ if the column exists. If the column ‘index_value’ does not exist, it will print a message. Please make sure the ‘index_value’ column exists and its name is spelled correctly.”,</blockquote>GPT-4.1 had an arguably even better solution. For nine of the 10 test cases, it simply printed the list of columns in the dataframe and included a comment in the code suggesting that I check to see if the column was present, and fix the issue if it wasn’t.GPT-5, by contrast, found a solution that worked every time: It simply took the actual index of each row (not the fictitious ‘index_value’) and added 1 to it in order to create new_column. This is the worst possible outcome: The code executes successfully, and at first glance seems to be doing the right thing, but the resulting value is essentially a random number. In a real-world example, this would create a much larger headache downstream in the code.<blockquote>df = pd.read_csv(‘data.csv’) df['new_column'] = df.index + 1</blockquote>I wondered if this issue was particular to the gpt family of models. I didn’t test every model in existence, but as a check, I repeated my experiment on Anthropic’s Claude models. I found the same trend: The older Claude models, confronted with this unsolvable problem, essentially shrug their shoulders, while the newer models sometimes solve the problem and sometimes just sweep it under the rug. <img alt="A chart showing the fraction of responses that were helpful, unhelpful, or counterproductive for different versions of large language models. " class="rm-shortcode" data-rm-shortcode-id="14fa8c844f8f1283d87dbf13b88a6f2d" data-rm-shortcode-name="rebelmouse-image" id="5be57" loading="lazy" src="https://spectrum.ieee.org/media-library/a-chart-showing-the-fraction-of-responses-that-were-helpful-unhelpful-or-counterproductive-for-different-versions-of-large-lan.jpg?id=62650386&width=980"/> Newer versions of large language models were more likely to produce counterproductive output when presented with a simple coding error. Jamie Twiss<h2>Garbage in, garbage out</h2>I don’t have inside knowledge on why the newer models fail in such a pernicious way. But I have an educated guess. I believe it’s the result of how the LLMs are being trained to code. The older models were trained on code much the same way they were trained on other text. Large volumes of presumably functional code were ingested as training data, which was used to set model weights. This wasn’t always perfect, as anyone using AI for coding in early 2023 will remember, with frequent syntax errors and faulty logic. But it certainly didn’t rip out safety checks or find ways to create plausible but fake data, like GPT-5 in my example above.But as soon as AI coding assistants arrived and were integrated into coding environments, the model creators realized they had a powerful source of labeled training data: the behavior of the users themselves. If an assistant offered up suggested code, the code ran successfully, and the user accepted the code, that was a positive signal, a sign that the assistant had gotten it right. If the user rejected the code, or if the code failed to run, that was a negative signal, and when the model was retrained, the assistant would be steered in a different direction.This is a powerful idea, and no doubt contributed to the rapid improvement of AI coding assistants for a period of time. But as inexperienced coders started turning up in greater numbers, it also started to poison the training data. AI coding assistants that found ways to get their code accepted by users kept doing more of that, even if “that” meant turning off safety checks and generating plausible but useless data. As long as a suggestion was taken on board, it was viewed as good, and downstream pain would be unlikely to be traced back to the source.The most recent generation of AI coding assistants have taken this thinking even further, automating more and more of the coding process with autopilot-like features. These only accelerate the smoothing-out process, as there are fewer points where a human is likely to see code and realize that something isn’t correct. Instead, the assistant is likely to keep iterating to try to get to a successful execution. In doing so, it is likely learning the wrong lessons.I am a huge believer in artificial intelligence, and I believe that AI coding assistants have a valuable role to play in accelerating development and democratizing the process of software creation. But chasing short-term gains, and relying on cheap, abundant, but ultimately poor-quality training data is going to continue resulting in model outcomes that are worse than useless. To start making models better again, AI coding companies need to invest in high-quality data, perhaps even paying experts to label AI-generated code. Otherwise, the models will continue to produce garbage, be trained on that garbage, and thereby produce even more garbage, eating their own tails.

Jan 7, 2026

Meet the IEEE Board-Nominated Candidates for President-Elect

<img src="https://spectrum.ieee.org/media-library/david-alan-koehler-and-manfred-schindler-smiling.jpg?id=62681591&width=1245&height=700&coordinates=0%2C156%2C0%2C157"/> The IEEE Board of Directors has nominated IEEE Senior Member David Alan Koehler and IEEE Life Fellow Manfred “Fred” J. Schindler as candidates for 2027 IEEE president-elect.IEEE Senior Member Gerardo Barbosa and IEEE Life Senior Member Timothy T. Lee are seeking nomination by petition. A separate article will be published in The Institute at a later date.The winner of this year’s election will serve as IEEE president in 2028. For more information about the election, president-elect candidates, and the petition process, visit the <a href="https://www.ieee.org/about/corporate/election/index.html" rel="noopener noreferrer" target="_blank">ieee.org/elections</a>.<h2>IEEE Senior Member David Alan Koehler</h2> <img alt="David Alan Koehler smiling in a suit jacket and tie." class="rm-shortcode" data-rm-shortcode-id="d47302ef50aa12d9f0c6de1e3c321c2a" data-rm-shortcode-name="rebelmouse-image" id="24785" loading="lazy" src="https://spectrum.ieee.org/media-library/david-alan-koehler-smiling-in-a-suit-jacket-and-tie.jpg?id=62681605&width=980"/> Steven Miller PhotographyKoehler is a subject matter expert with almost 30 years of experience in establishing condition-based maintenance practices for electrical equipment and managing analytical laboratories. He has presented his work at global conferences and published articles in technical publications related to the power industry. Koehler is an executive advisor at Danovo Energy Solutions.An active volunteer, he has served in every geographical unit within IEEE. His first leadership position was chair of the Central Indiana Section from 2012 to 2014. He served as 2019–2020 director of IEEE Region 4, vice chair of the 2022 IEEE Board of Directors Ad Hoc Committee on the Future of Engagement, 2022 vice president of IEEE Member and Geographic Activities, and chair of the 2024 IEEE Board of Directors Ad Hoc Committee on Leadership Continuity and Efficiency.He served on the IEEE Board of Directors for three different years. He has been a member of the IEEE-USA, Member and Geographic Activities, and Publication Services and Products boards.Koehler is a proud and active member of IEEE Women In Engineering and IEEE-Eta Kappa Nu, the honor society.<h2>IEEE Life Fellow Manfred “Fred” J. Schindler</h2> <img alt="Manfred Schindler smiling in a suit jacket and tie." class="rm-shortcode" data-rm-shortcode-id="5007f5dfdb37a665effc8b467b07fa9c" data-rm-shortcode-name="rebelmouse-image" id="ff563" loading="lazy" src="https://spectrum.ieee.org/media-library/manfred-schindler-smiling-in-a-suit-jacket-and-tie.jpg?id=62681616&width=980"/> Steven Miller PhotographySchindler, an expert in microwave semiconductor technology, is an independent consultant supporting clients with technical expertise, due diligence, and project management.Throughout his career, he led the development of microwave integrated-circuit technology, from lab demonstrations to high-volume commercial products. He has numerous technical publications and holds 11 patents.Schindler served as CTO of Anlotek, and director of Qorvo and RFMD’s Boston design center. He was applications manager at IBM, engineering manager at ATN Microwave, and a lab manager at Raytheon.An IEEE volunteer for more than 30 years, Schindler served as the 2024 vice president of IEEE Technical Activities and the 2022–2023 Division IV director. He was chair of the IEEE Conferences Committee from 2015 to 2018 and president of the IEEE Microwave Theory and Technology Society (MTTS) in 2003. He received the 2018 IEEE MTTS Distinguished Service Award. His award-winning micro-business column has appeared in IEEE Microwave Magazine since 2011.He also led the 2025 One IEEE to Enable Strategic Investments in Innovations and Public Imperative Activities adhoc committee.Schindler is an IEEE–Eta Kappa Nu honorary life member.

Jan 7, 2026

These Hearing Aids Will Tune in to Your Brain

<img src="https://spectrum.ieee.org/media-library/side-by-side-photos-show-a-woman-wearing-specialized-hearing-aids-each-with-a-cord-coming-down-from-it-and-a-close-up-of-the-d.jpg?id=62620525&width=1245&height=700&coordinates=0%2C160%2C0%2C161"/> Imagine you’re at a bustling dinner party filled with laughter, music, and clinking silverware. You’re trying to follow a conversation across the table, but every word feels like it’s wrapped in noise. For most people, these types of party scenarios, where it’s difficult to filter out extraneous sounds and focus on a single source, are an occasional annoyance. For millions with <a href="https://spectrum.ieee.org/tag/hearing-loss" target="_blank">hearing loss</a>, they’re a daily challenge—and not just in busy settings.Today’s <a href="https://spectrum.ieee.org/tag/hearing-aids" target="_blank">hearing aids</a> aren’t great at determining which sounds to amplify and which to ignore, and this often leaves users overwhelmed and fatigued. Even the routine act of conversing with a loved one during a car ride can be mentally draining, simply because the hum of the engine and road noises are magnified to create loud and constant background static that blurs speech.In recent years, modern hearing aids have made impressive strides. They can, for example, use a technology called adaptive beamforming to focus their microphones in the direction of a talker. Noise-reduction settings also help decrease background cacophony, and some devices even use machine-learning-based analysis, trained on uploaded data, to detect certain environments—for example a car or a party—and deploy custom settings.That’s why I was initially surprised to find out that today’s state-of-the-art hearing aids aren’t good enough. “It’s like my ears work but my brain is tired,” I remember one elderly man complaining, frustrated with the inadequacy of his cutting-edge noise-suppression hearing aids. At the time, I was a graduate student at the University of Texas at Dallas, surveying individuals with hearing loss. The man’s insight led me to a realization: Mental strain is an unaddressed frontier of hearing technology.But what if hearing aids were more than just amplifiers? What if they were listeners too? I envision a new generation of intelligent hearing aids that not only boost sound but also read the wearer’s brain waves and other key physiological markers, enabling them to react accordingly to improve hearing and counter fatigue.Until last spring, when I took time off to care for my child, I was a senior audio research scientist at <a href="https://www.harman.com/" target="_blank">Harman International</a>, in Los Angeles. My work combined cognitive neuroscience, auditory prosthetics, and the processing of biosignals, which are measurable physiological cues that reflect our mental and physical state. I’m passionate about developing brain-computer interfaces (<a href="https://spectrum.ieee.org/tag/bci" target="_blank">BCIs</a>) and adaptive signal-processing systems that make life easier for people with hearing loss. And I’m not alone. A number of researchers and companies are working to create smart hearing aids, and it’s likely they’ll come on the market within a decade.Two technologies in particular are poised to revolutionize hearing aids, offering personalized, fatigue-free listening experiences: electroencephalography (EEG), which tracks brain activity, and pupillometry, which uses eye measurements to gauge cognitive effort. These approaches might even be used to improve consumer audio devices, transforming the way we listen everywhere.<h2>Aging Populations in a Noisy World </h2>More than <a href="https://www.who.int/news-room/fact-sheets/detail/deafness-and-hearing-loss" rel="noopener noreferrer" target="_blank">430 million people</a> suffer from disabling hearing loss worldwide, including 34 million children, according to the World Health Organization. And the problem will likely get worse due to rising life expectancies and the fact that the world itself seems to be getting louder. By 2050, an estimated <a href="https://pubmed.ncbi.nlm.nih.gov/33714390/" target="_blank">2.5 billion people</a> will suffer some degree of hearing loss and 700 million will require intervention. On top of that, <a href="https://gh.bmj.com/content/7/11/e010501" rel="noopener noreferrer" target="_blank">as many as 1.4 billion of today’s young people</a>—nearly half of those aged 12 to 34—could be at risk of permanent hearing loss from listening to audio devices too loud and for too long.Every year, close to a trillion dollars is lost globally due to unaddressed hearing loss, a trend that is also likely getting more pronounced. That doesn’t account for the significant emotional and physical toll on the hearing impaired, including isolation, loneliness, depression, shame, anxiety, sleep disturbances, and loss of balance. <img alt="A back view of a man's head shows a flexible pattern of lines with electrodes inside that go over his ear and extend toward the front of his face." class="rm-shortcode" data-rm-shortcode-id="12fa3c83e74db09134be9565e28904d6" data-rm-shortcode-name="rebelmouse-image" id="83e0c" loading="lazy" src="https://spectrum.ieee.org/media-library/a-back-view-of-a-man-s-head-shows-a-flexible-pattern-of-lines-with-electrodes-inside-that-go-over-his-ear-and-extend-toward-the.jpg?id=62620547&width=980"/> Flex-printed electrode arrays, such as these from the Fraunhofer Institute for Digital Media Technology, offer a comfortable option for collecting high-quality EEG signals. Leona Hofmann/Fraunhofer IDMTAnd yet, despite widespread availability, hearing aid adoption remains low. According to a <a href="https://www.thelancet.com/journals/lanhl/article/PIIS2666-7568%2823%2900232-5/fulltext" target="_blank">2024 study</a> published in <a href="https://www.thelancet.com/" target="_blank">The Lancet,</a> only about 13 percent of American adults with hearing loss regularly wear hearing aids. Key reasons for this deficiency include discomfort, stigma, cost—and, crucially, frustration with the poor performance of hearing aids in noisy environments.Historically, hearing technology has come a long way. As early as the 13th century, people began using horns of cows and rams as “<a href="https://en.wikipedia.org/wiki/Ear_trumpet" target="_blank">ear trumpets</a>.” Commercial versions made of various materials, including brass and wood, came on the market in the early 19th century. (Beethoven, who famously began losing his hearing in his twenties, used variously shaped ear trumpets, some of which are now on display in a museum in Bonn, Germany.) But these contraptions were so bulky that users had to hold them with their hands or wear them within headbands. To avoid stigma, some even hid hearing aids inside furniture to mask their disability. In 1819, a <a href="https://www.bbc.com/news/blogs-ouch-29896747" target="_blank">special acoustic chair</a> was designed for the king of Portugal, featuring arms ornately carved to look like open lion mouths, which helped transmit sound to the king’s ear via speaking tubes.Modern hearing aids came into being after the advent of electronics in the early 20th century. Early devices used vacuum tubes and then transistors to amplify sound, shrinking over time from bulky body-worn boxes to discreet units that fit behind or inside the ear. At their core, today’s hearing aids still work on the same principle: A microphone picks up sound, a processor amplifies and shapes it to match the user’s hearing loss, and a tiny speaker delivers the adjusted sound into the ear canal.Today’s best-in-class devices, like those from <a href="https://www.oticon.com/solutions/real-hearing-aids" target="_blank">Oticon</a>, <a href="https://www.phonak.com/en-us/hearing-devices/hearing-aids/audeo-lumity" target="_blank">Phonak</a>, and <a href="https://www.starkey.com/hearing-aids/genesis-artificial-intelligence-hearing-aids" target="_blank">Starkey</a>, have pioneered increasingly advanced technologies, including the aforementioned beamforming microphones, frequency lowering to better pick up high-pitched sounds and voices, and machine learning to recognize and adapt to specific environments. For example, the device may reduce amplification in a quiet room to avoid escalating background hums or else increase amplification in a noisy café to make speech more intelligible. Advances in the AI technique of deep learning, which relies on artificial neural networks to automatically recognize patterns, also hold enormous promise. Using context-aware algorithms, this technology can, for example, be used to help distinguish between speech and noise, predict and suppress unwanted clamor in real time, and attempt to clean up speech that is muffled or distorted.The problem? As of right now, consumer systems respond only to external acoustic environments and not to the internal cognitive state of the listener—which means they act on imperfect and incomplete information. So, what if hearing aids were more empathetic? What if they could sense when the listener’s brain feels tired or overwhelmed and automatically use that feedback to deploy advanced features?<h2>Using EEG to Augment Hearing Aids</h2>When it comes to creating intelligent hearing aids, there are two main challenges. The first is building convenient, power-efficient wearable devices that accurately detect brain states. The second, perhaps more difficult step is decoding feedback from the brain and using that information to help hearing aids adapt in real time to the listener’s cognitive state and auditory experience.Let’s start with <a href="https://spectrum.ieee.org/tag/eeg" target="_blank">EEG</a>. This century-old noninvasive technology uses electrodes placed on the scalp to measure the brain’s electrical activity through voltage fluctuations, which are recorded as “brain waves.” <img alt="A man with headphones sits in a lab in front of computers displaying information. Behind him through a doorway is seen another person sitting in front of a screen, wearing an EEG cap." class="rm-shortcode" data-rm-shortcode-id="19caacab1cb3ed292373b00bcba7dbc3" data-rm-shortcode-name="rebelmouse-image" id="b6cbf" loading="lazy" src="https://spectrum.ieee.org/media-library/a-man-with-headphones-sits-in-a-lab-in-front-of-computers-displaying-information-behind-him-through-a-doorway-is-seen-another-p.jpg?id=62620561&width=980"/> Brain-computer interfaces allow researchers to accurately determine a listener’s focus in multitalker environments. Here, professor Christopher Smalt works on an attention-decoding system at the MIT Lincoln Laboratory.MIT Lincoln LaboratoryClinically, EEG has long been applied for diagnosing epilepsy and sleep disorders, monitoring brain injuries, assessing hearing ability in infants and impaired individuals, and more. And while standard EEG requires conductive gel and bulky headsets, we now have versions that are far more portable and convenient. These breakthroughs have already allowed EEG to migrate from hospitals into the consumer tech spaces, driving everything from neurofeedback headbands to the BCIs in gaming and wellness apps that allow people to control devices with their minds.The <a href="https://uol.de/psychologie/abteilungen/ceegrid" target="_blank">cEEGrid project</a> at Oldenburg University, in Germany, positions lightweight adhesive electrodes around the ear to create a low-profile version. In Denmark, <a href="https://ece.au.dk/en/research/research-centres/center-for-ear-eeg/" target="_blank">Aarhus University’s Center for Ear-EEG</a> also has an ear-based EEG system designed for comfort and portability. While the signal-to-noise ratio is slightly lower compared to head-worn EEG, these ear-based systems have proven sufficiently accurate for gauging attention, listening effort, <a href="https://ieeexplore.ieee.org/document/8006311" target="_blank">hearing thresholds</a>, and <a href="https://doi.org/10.1088/1741-2552/ae00f3" target="_blank">speech tracking</a> in real time.For hearing aids, EEG technology can pick up brain-wave patterns that reveal how well a listener is following speech: When listeners are paying attention, their brain rhythms synchronize with the syllabic rhythms of discourse, essentially tracking the speaker’s cadence. By contrast, if the signal becomes weaker or less precise, it suggests the listener is struggling to comprehend and losing focus.During my own <a href="https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.927872/full" target="_blank">Ph.D. research</a>, I observed firsthand how real-time brain-wave patterns, picked up by EEG, can reflect the quality of a listener’s speech cognition. For example, when participants successfully homed in on a single talker in a crowded room, their neural rhythms aligned nearly perfectly with that speaker’s voice. It was as if there were a brain-based spotlight on that speaker! But when background fracas grew louder or the listener’s attention drifted, those patterns waned, revealing stress in keeping up.Today, researchers at <a href="https://ece.au.dk/en/research/research-centres/center-for-ear-eeg/projects/investigation-of-auditory-responses-in-ear-eeg?utm_source=chatgpt.com" target="_blank">Aarhus University</a>, <a href="https://uol.de/en/psychology/neurophysiology/forschung" target="_blank">Oldenburg University</a>, and <a href="https://www.ll.mit.edu/partner-us/available-technologies/end-end-deep-neural-network-auditory-attention-decoding?utm_source=chatgpt.com" target="_blank">MIT</a> are developing attention-decoding algorithms specifically for auditory applications. For example, Oldenburg’s cEEGrid technology has been used to <a href="https://iopscience.iop.org/article/10.1088/1741-2552/aa66dd" target="_blank">successfully identify</a> which of two speakers a listener is trying to hear. In a <a href="https://iopscience.iop.org/article/10.1088/1741-2560/13/6/066004" target="_blank">related study</a>, researchers demonstrated that in-ear EEG can track the attended speech stream in multitalker environments.All of this could prove transformational in creating neuroadaptive hearing aids. If a listener’s EEG reveals a drop in speech tracking, the hearing aid could infer increased listening difficulty, even if ambient noise levels have remained constant. For example, if a hearing-impaired car driver can’t focus on a conversation due to mental fatigue caused by background noise, the hearing aid could switch on beamforming to better augment the passenger’s voice, as well as machine-learning settings to deploy sound canceling that blocks the din of the road.Of course, there are several hurdles to cross before commercialization becomes possible. For one thing, EEG-paired hearing aids will need to handle the fact that neural responses differ from person to person, which means they will likely need to be calibrated individually to capture each user’s unique brain-speech patterns.Additionally, EEG signals are themselves notoriously “noisy,” especially in real-world environments. Luckily, we already have algorithms and processing tools for cleaning and organizing these signals so computer models can search for key patterns that predict mental states, including attention drift and fatigue.Commercial versions of EEG-paired hearing aids will also need to be small and energy-efficient when it comes to signal processing and real-time computation. And getting them to work reliably, despite head movement and daily activity, will be no small feat. Importantly, companies will need to resolve ethical and regulatory considerations, such as data ownership. To me, these challenges seem surmountable, especially with technology progressing at a rapid clip.<h2>A Window to the Brain: Using Our Eyes to Hear</h2>Now let’s consider a second way of reading brain states: through the listener’s eyes.When a person has trouble hearing and starts feeling overwhelmed, the body reacts. Heart-rate variability diminishes, indicating stress, and sweating increases. Researchers are investigating how these types of autonomic nervous-system responses can be measured and used to create smart hearing aids. For the purposes of this article, I will focus on a response that seems especially promising—namely, pupil size.<a href="https://en.wikipedia.org/wiki/Pupillometry" target="_blank">Pupillometry</a> is the measurement of pupil size and how it changes in response to stimuli. We all know that pupils expand or contract depending on light brightness. As it turns out, pupil size is also an accurate means of evaluating attention, arousal, mental strain—and, crucially, listening effort. <img alt="Three eye illustrations showing pupil size changes due to light and emotional stimuli." class="rm-shortcode" data-rm-shortcode-id="1e09afde9216a40d6eb56b4443439f2c" data-rm-shortcode-name="rebelmouse-image" id="8c8bd" loading="lazy" src="https://spectrum.ieee.org/media-library/three-eye-illustrations-showing-pupil-size-changes-due-to-light-and-emotional-stimuli.png?id=62712190&width=980"/> Pupil size is determined by both external stimuli, such as light, and internal stimuli, such as fatigue or excitement.Chris PhilpotIn recent years, studies at <a href="https://discovery.ucl.ac.uk/id/eprint/10173304/1/4856.full.pdf" target="_blank">University College London</a> and <a href="https://pubmed.ncbi.nlm.nih.gov/29435963/" target="_blank">Leiden University</a>, in the Netherlands, have demonstrated that pupil dilation is consistently greater in hearing-impaired individuals when processing speech in noisy conditions. Research has also shown pupillometry to be a sensitive, objective correlate of speech intelligibility and mental strain. It could therefore offer a feedback mechanism for user-aware hearing aids that dynamically adjust amplification strategies, directional focus, or noise reduction based not just on the acoustic environment but on how hard the user is working to comprehend speech.While more straightforward than EEG, pupillometry presents its own engineering challenges. Pupillometry requires a direct line of sight to the pupil, necessitating a stable, front-facing camera-to-eye configuration—which isn’t easy to achieve when a wearer is moving around in real-world settings. On top of that, most pupil-tracking systems require infrared illumination and high-resolution optical cameras, which are too bulky and power intensive for the tiny housings of in-ear or behind-the-ear hearing aids. All this makes it unlikely that standalone hearing aids will include pupil-tracking hardware in the near future.A more viable approach may be pairing hearing aids with smart glasses or other wearables that contain the necessary eye-tracking hardware. Products from companies like <a href="https://www.tobii.com/" target="_blank">Tobii</a> and <a href="https://pupil-labs.com/" target="_blank">Pupil Labs</a> already offer real-time pupillometry via lightweight headgear for use in research, behavioral analysis, and assistive technology for people with medical conditions that limit movement but leave eye control intact. Apple’s Vision Pro and other augmented reality or virtual reality platforms also include built-in eye-tracking sensors that could support pupillometry-driven adaptations for audio content. <img alt="A woman wears a pair of specialized glasses that have small cameras and infrared illuminators around edges of the glass for eye tracking, as well as a camera and microphone above the nose bridge." class="rm-shortcode" data-rm-shortcode-id="61ffe5f11715a80f11deca4c2cf9149d" data-rm-shortcode-name="rebelmouse-image" id="061a4" loading="lazy" src="https://spectrum.ieee.org/media-library/a-woman-wears-a-pair-of-specialized-glasses-that-have-small-cameras-and-infrared-illuminators-around-edges-of-the-glass-for-eye.jpg?id=62620574&width=980"/> Smart glasses that measure pupil size, such as these made by Tobii, could help determine listening strain. TobiiOnce pupil data is acquired, the next step will be real-time interpretation. Here, again, is where machine learning can use large datasets to detect patterns signifying increased cognitive load or attentional shifts. For instance, if a listener’s pupils dilate unnaturally during a conversation, signifying strain, the hearing aid could automatically engage a more aggressive noise suppression mode or narrow its directional microphone beam. These types of systems can also learn from contextual features, such as time of day or prior environments, to continuously refine their response strategies.While no commercial hearing aid currently integrates pupillometry, adjacent industries are moving quickly. <a href="https://emteqlabs.com/" target="_blank">Emteq Labs</a> is developing “emotion-sensing” glasses that combine facial and eye tracking, along with pupil measurement, to do things like evaluate mental health and capture consumer insights. Ethical controversies aside—just imagine what dystopian governments might do with emotion-reading eyewear!—such devices show that it’s feasible to embed biosignal monitoring in consumer-grade smart glasses.<h2>A Future with Empathetic Hearing Aids</h2>Back at the dinner party, it remains nearly impossible to participate in conversation. “Why even bother going out?” some ask. But that will soon change.We’re at the cusp of a paradigm shift in auditory technology, from device-centered to user-centered innovation. In the next five years, we may see hybrid solutions where EEG-enabled earbuds work in tandem with smart glasses. In 10 years, fully integrated biosignal-driven hearing aids could become the standard. And in 50? Perhaps audio systems will evolve into cognitive companions, devices that adjust, advise, and align with our mental state.Personalizing hearing-assistance technology isn’t just about improving clarity; it’s also about easing mental fatigue, reducing social isolation, and empowering people to engage confidently with the world. Ultimately, it’s about restoring dignity, connection, and joy.

Feb 9, 2022

Andrew Ng: Unbiggen AI

<img src="https://spectrum.ieee.org/media-library/andrew-ng-listens-during-the-power-of-data-sooner-than-you-think-global-technology-conference-in-brooklyn-new-york-on-wednes.jpg?id=29206806&width=1245&height=700&coordinates=0%2C0%2C0%2C474"/> <a href="https://en.wikipedia.org/wiki/Andrew_Ng" rel="noopener noreferrer" target="_blank">Andrew Ng</a> has serious street cred in artificial intelligence. He pioneered the use of graphics processing units (GPUs) to train deep learning models in the late 2000s with his students at <a href="https://stanfordmlgroup.github.io/" rel="noopener noreferrer" target="_blank">Stanford University</a>, cofounded <a href="https://research.google/teams/brain/" rel="noopener noreferrer" target="_blank">Google Brain</a> in 2011, and then served for three years as chief scientist for <a href="https://ir.baidu.com/" rel="noopener noreferrer" target="_blank">Baidu</a>, where he helped build the Chinese tech giant’s AI group. So when he says he has identified the next big shift in artificial intelligence, people listen. And that’s what he told IEEE Spectrum in an exclusive Q&A.<hr/> Ng’s current efforts are focused on his company <a href="https://landing.ai/about/" rel="noopener noreferrer" target="_blank">Landing AI</a>, which built a platform called LandingLens to help manufacturers improve visual inspection with computer vision. He has also become something of an evangelist for what he calls the <a href="https://www.youtube.com/watch?v=06-AZXmwHjo" target="_blank">data-centric AI movement</a>, which he says can yield “small data” solutions to big issues in AI, including model efficiency, accuracy, and bias. Andrew Ng on... <ul> <li><a href="#big">What’s next for really big models</a></li> <li><a href="#career">The career advice he didn’t listen to</a></li> <li><a href="#defining">Defining the data-centric AI movement</a></li> <li><a href="#synthetic">Synthetic data</a></li> <li><a href="#work">Why Landing AI asks its customers to do the work</a></li> </ul> The great advances in deep learning over the past decade or so have been powered by ever-bigger models crunching ever-bigger amounts of data. Some people argue that that’s an <a href="https://spectrum.ieee.org/deep-learning-computational-cost" target="_self">unsustainable trajectory</a>. Do you agree that it can’t go on that way? Andrew Ng: This is a big question. We’ve seen foundation models in NLP [natural language processing]. I’m excited about NLP models getting even bigger, and also about the potential of building foundation models in computer vision. I think there’s lots of signal to still be exploited in video: We have not been able to build foundation models yet for video because of compute bandwidth and the cost of processing video, as opposed to tokenized text. So I think that this engine of scaling up deep learning algorithms, which has been running for something like 15 years now, still has steam in it. Having said that, it only applies to certain problems, and there’s a set of other problems that need small data solutions. When you say you want a foundation model for computer vision, what do you mean by that? Ng: This is a term coined by <a href="https://cs.stanford.edu/~pliang/" rel="noopener noreferrer" target="_blank">Percy Liang</a> and <a href="https://crfm.stanford.edu/" rel="noopener noreferrer" target="_blank">some of my friends at Stanford</a> to refer to very large models, trained on very large data sets, that can be tuned for specific applications. For example, <a href="https://spectrum.ieee.org/open-ais-powerful-text-generating-tool-is-ready-for-business" target="_self">GPT-3</a> is an example of a foundation model [for NLP]. Foundation models offer a lot of promise as a new paradigm in developing machine learning applications, but also challenges in terms of making sure that they’re reasonably fair and free from bias, especially if many of us will be building on top of them. What needs to happen for someone to build a foundation model for video? Ng: I think there is a scalability problem. The compute power needed to process the large volume of images for video is significant, and I think that’s why foundation models have arisen first in NLP. Many researchers are working on this, and I think we’re seeing early signs of such models being developed in computer vision. But I’m confident that if a semiconductor maker gave us 10 times more processor power, we could easily find 10 times more video to build such models for vision. Having said that, a lot of what’s happened over the past decade is that deep learning has happened in consumer-facing companies that have large user bases, sometimes billions of users, and therefore very large data sets. While that paradigm of machine learning has driven a lot of economic value in consumer software, I find that that recipe of scale doesn’t work for other industries. <a href="#top">Back to top</a> It’s funny to hear you say that, because your early work was at a consumer-facing company with millions of users. Ng: Over a decade ago, when I proposed starting the <a href="https://research.google/teams/brain/" rel="noopener noreferrer" target="_blank">Google Brain</a> project to use Google’s compute infrastructure to build very large neural networks, it was a controversial step. One very senior person pulled me aside and warned me that starting Google Brain would be bad for my career. I think he felt that the action couldn’t just be in scaling up, and that I should instead focus on architecture innovation. “In many industries where giant data sets simply don’t exist, I think the focus has to shift from big data to good data. Having 50 thoughtfully engineered examples can be sufficient to explain to the neural network what you want it to learn.” —Andrew Ng, CEO & Founder, Landing AI I remember when my students and I published the first <a href="https://nips.cc/" rel="noopener noreferrer" target="_blank">NeurIPS</a> workshop paper advocating using <a href="https://developer.nvidia.com/cuda-zone" rel="noopener noreferrer" target="_blank">CUDA</a>, a platform for processing on GPUs, for deep learning—a different senior person in AI sat me down and said, “CUDA is really complicated to program. As a programming paradigm, this seems like too much work.” I did manage to convince him; the other person I did not convince. I expect they’re both convinced now. Ng: I think so, yes. Over the past year as I’ve been speaking to people about the data-centric AI movement, I’ve been getting flashbacks to when I was speaking to people about deep learning and scalability 10 or 15 years ago. In the past year, I’ve been getting the same mix of “there’s nothing new here” and “this seems like the wrong direction.” <a href="#top">Back to top</a> How do you define data-centric AI, and why do you consider it a movement? Ng: Data-centric AI is the discipline of systematically engineering the data needed to successfully build an AI system. For an AI system, you have to implement some algorithm, say a neural network, in code and then train it on your data set. The dominant paradigm over the last decade was to download the data set while you focus on improving the code. Thanks to that paradigm, over the last decade deep learning networks have improved significantly, to the point where for a lot of applications the code—the neural network architecture—is basically a solved problem. So for many practical applications, it’s now more productive to hold the neural network architecture fixed, and instead find ways to improve the data. When I started speaking about this, there were many practitioners who, completely appropriately, raised their hands and said, “Yes, we’ve been doing this for 20 years.” This is the time to take the things that some individuals have been doing intuitively and make it a systematic engineering discipline. The data-centric AI movement is much bigger than one company or group of researchers. My collaborators and I organized a <a href="https://neurips.cc/virtual/2021/workshop/21860" rel="noopener noreferrer" target="_blank">data-centric AI workshop at NeurIPS</a>, and I was really delighted at the number of authors and presenters that showed up. You often talk about companies or institutions that have only a small amount of data to work with. How can data-centric AI help them? Ng: You hear a lot about vision systems built with millions of images—I once built a face recognition system using 350 million images. Architectures built for hundreds of millions of images don’t work with only 50 images. But it turns out, if you have 50 really good examples, you can build something valuable, like a defect-inspection system. In many industries where giant data sets simply don’t exist, I think the focus has to shift from big data to good data. Having 50 thoughtfully engineered examples can be sufficient to explain to the neural network what you want it to learn. When you talk about training a model with just 50 images, does that really mean you’re taking an existing model that was trained on a very large data set and fine-tuning it? Or do you mean a brand new model that’s designed to learn only from that small data set? Ng: Let me describe what Landing AI does. When doing visual inspection for manufacturers, we often use our own flavor of <a href="https://developers.arcgis.com/python/guide/how-retinanet-works/" rel="noopener noreferrer" target="_blank">RetinaNet</a>. It is a pretrained model. Having said that, the pretraining is a small piece of the puzzle. What’s a bigger piece of the puzzle is providing tools that enable the manufacturer to pick the right set of images [to use for fine-tuning] and label them in a consistent way. There’s a very practical problem we’ve seen spanning vision, NLP, and speech, where even human annotators don’t agree on the appropriate label. For big data applications, the common response has been: If the data is noisy, let’s just get a lot of data and the algorithm will average over it. But if you can develop tools that flag where the data’s inconsistent and give you a very targeted way to improve the consistency of the data, that turns out to be a more efficient way to get a high-performing system. “Collecting more data often helps, but if you try to collect more data for everything, that can be a very expensive activity.” —Andrew Ng For example, if you have 10,000 images where 30 images are of one class, and those 30 images are labeled inconsistently, one of the things we do is build tools to draw your attention to the subset of data that’s inconsistent. So you can very quickly relabel those images to be more consistent, and this leads to improvement in performance. Could this focus on high-quality data help with bias in data sets? If you’re able to curate the data more before training? Ng: Very much so. Many researchers have pointed out that biased data is one factor among many leading to biased systems. There have been many thoughtful efforts to engineer the data. At the NeurIPS workshop, <a href="https://www.cs.princeton.edu/~olgarus/" rel="noopener noreferrer" target="_blank">Olga Russakovsky</a> gave a really nice talk on this. At the main NeurIPS conference, I also really enjoyed <a href="https://neurips.cc/virtual/2021/invited-talk/22281" rel="noopener noreferrer" target="_blank">Mary Gray’s presentation,</a> which touched on how data-centric AI is one piece of the solution, but not the entire solution. New tools like <a href="https://www.microsoft.com/en-us/research/project/datasheets-for-datasets/" rel="noopener noreferrer" target="_blank">Datasheets for Datasets</a> also seem like an important piece of the puzzle. One of the powerful tools that data-centric AI gives us is the ability to engineer a subset of the data. Imagine training a machine-learning system and finding that its performance is okay for most of the data set, but its performance is biased for just a subset of the data. If you try to change the whole neural network architecture to improve the performance on just that subset, it’s quite difficult. But if you can engineer a subset of the data you can address the problem in a much more targeted way. When you talk about engineering the data, what do you mean exactly? Ng: In AI, data cleaning is important, but the way the data has been cleaned has often been in very manual ways. In computer vision, someone may visualize images through a <a href="https://jupyter.org/" rel="noopener noreferrer" target="_blank">Jupyter notebook</a> and maybe spot the problem, and maybe fix it. But I’m excited about tools that allow you to have a very large data set, tools that draw your attention quickly and efficiently to the subset of data where, say, the labels are noisy. Or to quickly bring your attention to the one class among 100 classes where it would benefit you to collect more data. Collecting more data often helps, but if you try to collect more data for everything, that can be a very expensive activity. For example, I once figured out that a speech-recognition system was performing poorly when there was car noise in the background. Knowing that allowed me to collect more data with car noise in the background, rather than trying to collect more data for everything, which would have been expensive and slow. <a href="#top">Back to top</a> What about using synthetic data, is that often a good solution? Ng: I think synthetic data is an important tool in the tool chest of data-centric AI. At the NeurIPS workshop, <a href="https://tensorlab.cms.caltech.edu/users/anima/" rel="noopener noreferrer" target="_blank">Anima Anandkumar</a> gave a great talk that touched on synthetic data. I think there are important uses of synthetic data that go beyond just being a preprocessing step for increasing the data set for a learning algorithm. I’d love to see more tools to let developers use synthetic data generation as part of the closed loop of iterative machine learning development. Do you mean that synthetic data would allow you to try the model on more data sets? Ng: Not really. Here’s an example. Let’s say you’re trying to detect defects in a smartphone casing. There are many different types of defects on smartphones. It could be a scratch, a dent, pit marks, discoloration of the material, other types of blemishes. If you train the model and then find through error analysis that it’s doing well overall but it’s performing poorly on pit marks, then synthetic data generation allows you to address the problem in a more targeted way. You could generate more data just for the pit-mark category. “In the consumer software Internet, we could train a handful of machine-learning models to serve a billion users. In manufacturing, you might have 10,000 manufacturers building 10,000 custom AI models.” —Andrew Ng Synthetic data generation is a very powerful tool, but there are many simpler tools that I will often try first. Such as data augmentation, improving labeling consistency, or just asking a factory to collect more data. <a href="#top">Back to top</a> To make these issues more concrete, can you walk me through an example? When a company approaches <a href="https://landing.ai/" rel="noopener noreferrer" target="_blank">Landing AI</a> and says it has a problem with visual inspection, how do you onboard them and work toward deployment? Ng: When a customer approaches us we usually have a conversation about their inspection problem and look at a few images to verify that the problem is feasible with computer vision. Assuming it is, we ask them to upload the data to the <a href="https://landing.ai/platform/" rel="noopener noreferrer" target="_blank">LandingLens</a> platform. We often advise them on the methodology of data-centric AI and help them label the data. One of the foci of Landing AI is to empower manufacturing companies to do the machine learning work themselves. A lot of our work is making sure the software is fast and easy to use. Through the iterative process of machine learning development, we advise customers on things like how to train models on the platform, when and how to improve the labeling of data so the performance of the model improves. Our training and software supports them all the way through deploying the trained model to an edge device in the factory. How do you deal with changing needs? If products change or lighting conditions change in the factory, can the model keep up? Ng: It varies by manufacturer. There is data drift in many contexts. But there are some manufacturers that have been running the same manufacturing line for 20 years now with few changes, so they don’t expect changes in the next five years. Those stable environments make things easier. For other manufacturers, we provide tools to flag when there’s a significant data-drift issue. I find it really important to empower manufacturing customers to correct data, retrain, and update the model. Because if something changes and it’s 3 a.m. in the United States, I want them to be able to adapt their learning algorithm right away to maintain operations. In the consumer software Internet, we could train a handful of machine-learning models to serve a billion users. In manufacturing, you might have 10,000 manufacturers building 10,000 custom AI models. The challenge is, how do you do that without Landing AI having to hire 10,000 machine learning specialists? So you’re saying that to make it scale, you have to empower customers to do a lot of the training and other work. Ng: Yes, exactly! This is an industry-wide problem in AI, not just in manufacturing. Look at health care. Every hospital has its own slightly different format for electronic health records. How can every hospital train its own custom AI model? Expecting every hospital’s IT personnel to invent new neural-network architectures is unrealistic. The only way out of this dilemma is to build tools that empower the customers to build their own models by giving them tools to engineer the data and express their domain knowledge. That’s what Landing AI is executing in computer vision, and the field of AI needs other teams to execute this in other domains. Is there anything else you think it’s important for people to understand about the work you’re doing or the data-centric AI movement? Ng: In the last decade, the biggest shift in AI was a shift to deep learning. I think it’s quite possible that in this decade the biggest shift will be to data-centric AI. With the maturity of today’s neural network architectures, I think for a lot of the practical applications the bottleneck will be whether we can efficiently get the data we need to develop systems that work well. The data-centric AI movement has tremendous energy and momentum across the whole community. I hope more researchers and developers will jump in and work on it. <a href="#top">Back to top</a> This article appears in the April 2022 print issue as “Andrew Ng, AI Minimalist.”

Feb 8, 2022

How AI Will Change Chip Design

<img src="https://spectrum.ieee.org/media-library/layered-rendering-of-colorful-semiconductor-wafers-with-a-bright-white-light-sitting-on-one.jpg?id=29285079&width=1245&height=700&coordinates=0%2C156%2C0%2C156"/> The end of <a href="https://spectrum.ieee.org/on-beyond-moores-law-4-new-laws-of-computing" target="_self">Moore’s Law</a> is looming. Engineers and designers can do only so much to <a href="https://spectrum.ieee.org/ibm-introduces-the-worlds-first-2nm-node-chip" target="_self">miniaturize transistors</a> and <a href="https://spectrum.ieee.org/cerebras-giant-ai-chip-now-has-a-trillions-more-transistors" target="_self">pack as many of them as possible into chips</a>. So they’re turning to other approaches to chip design, incorporating technologies like AI into the process.Samsung, for instance, is <a href="https://spectrum.ieee.org/processing-in-dram-accelerates-ai" target="_self">adding AI to its memory chips</a> to enable processing in memory, thereby saving energy and speeding up machine learning. Speaking of speed, Google’s TPU V4 AI chip has <a href="https://spectrum.ieee.org/heres-how-googles-tpu-v4-ai-chip-stacked-up-in-training-tests" target="_self">doubled its processing power</a> compared with that of its previous version.But AI holds still more promise and potential for the semiconductor industry. To better understand how AI is set to revolutionize chip design, we spoke with <a href="https://www.linkedin.com/in/heather-gorr-phd" rel="noopener noreferrer" target="_blank">Heather Gorr</a>, senior product manager for <a href="https://www.mathworks.com/" rel="noopener noreferrer" target="_blank">MathWorks</a>’ MATLAB platform.How is AI currently being used to design the next generation of chips?Heather Gorr: AI is such an important technology because it’s involved in most parts of the cycle, including the design and manufacturing process. There’s a lot of important applications here, even in the general process engineering where we want to optimize things. I think defect detection is a big one at all phases of the process, especially in manufacturing. But even thinking ahead in the design process, [AI now plays a significant role] when you’re designing the light and the sensors and all the different components. There’s a lot of anomaly detection and fault mitigation that you really want to consider. <img alt="Portrait of a woman with blonde-red hair smiling at the camera" class="rm-shortcode rm-resized-image" data-rm-shortcode-id="1f18a02ccaf51f5c766af2ebc4af18e1" data-rm-shortcode-name="rebelmouse-image" id="2dc00" loading="lazy" src="https://spectrum.ieee.org/media-library/portrait-of-a-woman-with-blonde-red-hair-smiling-at-the-camera.jpg?id=29288554&width=980" style="max-width: 100%"/> Heather GorrMathWorksThen, thinking about the logistical modeling that you see in any industry, there is always planned downtime that you want to mitigate; but you also end up having unplanned downtime. So, looking back at that historical data of when you’ve had those moments where maybe it took a bit longer than expected to manufacture something, you can take a look at all of that data and use AI to try to identify the proximate cause or to see something that might jump out even in the processing and design phases. We think of AI oftentimes as a predictive tool, or as a robot doing something, but a lot of times you get a lot of insight from the data through AI.What are the benefits of using AI for chip design?Gorr: Historically, we’ve seen a lot of physics-based modeling, which is a very intensive process. We want to do a <a href="https://en.wikipedia.org/wiki/Model_order_reduction" rel="noopener noreferrer" target="_blank">reduced order model</a>, where instead of solving such a computationally expensive and extensive model, we can do something a little cheaper. You could create a surrogate model, so to speak, of that physics-based model, use the data, and then do your <a href="https://institutefordiseasemodeling.github.io/idmtools/parameter-sweeps.html" rel="noopener noreferrer" target="_blank">parameter sweeps</a>, your optimizations, your <a href="https://www.ibm.com/cloud/learn/monte-carlo-simulation" rel="noopener noreferrer" target="_blank">Monte Carlo simulations</a> using the surrogate model. That takes a lot less time computationally than solving the physics-based equations directly. So, we’re seeing that benefit in many ways, including the efficiency and economy that are the results of iterating quickly on the experiments and the simulations that will really help in the design.So it’s like having a digital twin in a sense?Gorr: Exactly. That’s pretty much what people are doing, where you have the physical system model and the experimental data. Then, in conjunction, you have this other model that you could tweak and tune and try different parameters and experiments that let sweep through all of those different situations and come up with a better design in the end.So, it’s going to be more efficient and, as you said, cheaper?Gorr: Yeah, definitely. Especially in the experimentation and design phases, where you’re trying different things. That’s obviously going to yield dramatic cost savings if you’re actually manufacturing and producing [the chips]. You want to simulate, test, experiment as much as possible without making something using the actual process engineering.We’ve talked about the benefits. How about the drawbacks?Gorr: The [AI-based experimental models] tend to not be as accurate as physics-based models. Of course, that’s why you do many simulations and parameter sweeps. But that’s also the benefit of having that digital twin, where you can keep that in mind—it’s not going to be as accurate as that precise model that we’ve developed over the years.Both chip design and manufacturing are system intensive; you have to consider every little part. And that can be really challenging. It’s a case where you might have models to predict something and different parts of it, but you still need to bring it all together.One of the other things to think about too is that you need the data to build the models. You have to incorporate data from all sorts of different sensors and different sorts of teams, and so that heightens the challenge.How can engineers use AI to better prepare and extract insights from hardware or sensor data?Gorr: We always think about using AI to predict something or do some robot task, but you can use AI to come up with patterns and pick out things you might not have noticed before on your own. People will use AI when they have high-frequency data coming from many different sensors, and a lot of times it’s useful to explore the frequency domain and things like data synchronization or resampling. Those can be really challenging if you’re not sure where to start.One of the things I would say is, use the tools that are available. There’s a vast community of people working on these things, and you can find lots of examples [of applications and techniques] on <a href="https://github.com/" rel="noopener noreferrer" target="_blank">GitHub</a> or <a href="https://www.mathworks.com/matlabcentral/" rel="noopener noreferrer" target="_blank">MATLAB Central</a>, where people have shared nice examples, even little apps they’ve created. I think many of us are buried in data and just not sure what to do with it, so definitely take advantage of what’s already out there in the community. You can explore and see what makes sense to you, and bring in that balance of domain knowledge and the insight you get from the tools and AI.What should engineers and designers consider when using AI for chip design?Gorr: Think through what problems you’re trying to solve or what insights you might hope to find, and try to be clear about that. Consider all of the different components, and document and test each of those different parts. Consider all of the people involved, and explain and hand off in a way that is sensible for the whole team.How do you think AI will affect chip designers’ jobs?Gorr: It’s going to free up a lot of human capital for more advanced tasks. We can use AI to reduce waste, to optimize the materials, to optimize the design, but then you still have that human involved whenever it comes to decision-making. I think it’s a great example of people and technology working hand in hand. It’s also an industry where all people involved—even on the manufacturing floor—need to have some level of understanding of what’s happening, so this is a great industry for advancing AI because of how we test things and how we think about them before we put them on the chip.How do you envision the future of AI and chip design?Gorr: It’s very much dependent on that human element—involving people in the process and having that interpretable model. We can do many things with the mathematical minutiae of modeling, but it comes down to how people are using it, how everybody in the process is understanding and applying it. Communication and involvement of people of all skill levels in the process are going to be really important. We’re going to see less of those superprecise predictions and more transparency of information, sharing, and that digital twin—not only using AI but also using our human knowledge and all of the work that many people have done over the years.

Feb 7, 2022

Atomically Thin Materials Significantly Shrink Qubits

<img src="https://spectrum.ieee.org/media-library/a-golden-square-package-holds-a-small-processor-sitting-on-top-is-a-metal-square-with-mit-etched-into-it.jpg?id=29281587&width=1245&height=700&coordinates=0%2C156%2C0%2C156"/> Quantum computing is a devilishly complex technology, with many technical hurdles impacting its development. Of these challenges two critical issues stand out: miniaturization and qubit quality.IBM has adopted the superconducting qubit road map of <a href="https://spectrum.ieee.org/ibms-envisons-the-road-to-quantum-computing-like-an-apollo-mission" target="_self">reaching a 1,121-qubit processor by 2023</a>, leading to the expectation that 1,000 qubits with today’s qubit form factor is feasible. However, current approaches will require very large chips (50 millimeters on a side, or larger) at the scale of small wafers, or the use of chiplets on multichip modules. While this approach will work, the aim is to attain a better path toward scalability.Now researchers at <a href="https://www.nature.com/articles/s41563-021-01187-w" rel="noopener noreferrer" target="_blank">MIT have been able to both reduce the size of the qubits</a> and done so in a way that reduces the interference that occurs between neighboring qubits. The MIT researchers have increased the number of superconducting qubits that can be added onto a device by a factor of 100.“We are addressing both qubit miniaturization and quality,” said <a href="https://equs.mit.edu/william-d-oliver/" rel="noopener noreferrer" target="_blank">William Oliver</a>, the director for the <a href="https://cqe.mit.edu/" target="_blank">Center for Quantum Engineering</a> at MIT. “Unlike conventional transistor scaling, where only the number really matters, for qubits, large numbers are not sufficient, they must also be high-performance. Sacrificing performance for qubit number is not a useful trade in quantum computing. They must go hand in hand.”The key to this big increase in qubit density and reduction of interference comes down to the use of two-dimensional materials, in particular the 2D insulator hexagonal boron nitride (hBN). The MIT researchers demonstrated that a few atomic monolayers of hBN can be stacked to form the insulator in the capacitors of a superconducting qubit.Just like other capacitors, the capacitors in these superconducting circuits take the form of a sandwich in which an insulator material is sandwiched between two metal plates. The big difference for these capacitors is that the superconducting circuits can operate only at extremely low temperatures—less than 0.02 degrees above absolute zero (-273.15 °C). <img alt="Golden dilution refrigerator hanging vertically" class="rm-shortcode rm-resized-image" data-rm-shortcode-id="694399af8a1c345e51a695ff73909eda" data-rm-shortcode-name="rebelmouse-image" id="6c615" loading="lazy" src="https://spectrum.ieee.org/media-library/golden-dilution-refrigerator-hanging-vertically.jpg?id=29281593&width=980" style="max-width: 100%"/> Superconducting qubits are measured at temperatures as low as 20 millikelvin in a dilution refrigerator.Nathan Fiske/MITIn that environment, insulating materials that are available for the job, such as PE-CVD silicon oxide or silicon nitride, have quite a few defects that are too lossy for quantum computing applications. To get around these material shortcomings, most superconducting circuits use what are called coplanar capacitors. In these capacitors, the plates are positioned laterally to one another, rather than on top of one another.As a result, the intrinsic silicon substrate below the plates and to a smaller degree the vacuum above the plates serve as the capacitor dielectric. Intrinsic silicon is chemically pure and therefore has few defects, and the large size dilutes the electric field at the plate interfaces, all of which leads to a low-loss capacitor. The lateral size of each plate in this open-face design ends up being quite large (typically 100 by 100 micrometers) in order to achieve the required capacitance.In an effort to move away from the large lateral configuration, the MIT researchers embarked on a search for an insulator that has very few defects and is compatible with superconducting capacitor plates.“We chose to study hBN because it is the most widely used insulator in 2D material research due to its cleanliness and chemical inertness,” said colead author <a href="https://equs.mit.edu/joel-wang/" rel="noopener noreferrer" target="_blank">Joel Wang</a>, a research scientist in the Engineering Quantum Systems group of the MIT Research Laboratory for Electronics. On either side of the hBN, the MIT researchers used the 2D superconducting material, niobium diselenide. One of the trickiest aspects of fabricating the capacitors was working with the niobium diselenide, which oxidizes in seconds when exposed to air, according to Wang. This necessitates that the assembly of the capacitor occur in a glove box filled with argon gas.While this would seemingly complicate the scaling up of the production of these capacitors, Wang doesn’t regard this as a limiting factor.“What determines the quality factor of the capacitor are the two interfaces between the two materials,” said Wang. “Once the sandwich is made, the two interfaces are “sealed” and we don’t see any noticeable degradation over time when exposed to the atmosphere.”This lack of degradation is because around 90 percent of the electric field is contained within the sandwich structure, so the oxidation of the outer surface of the niobium diselenide does not play a significant role anymore. This ultimately makes the capacitor footprint much smaller, and it accounts for the reduction in cross talk between the neighboring qubits.“The main challenge for scaling up the fabrication will be the wafer-scale growth of hBN and 2D superconductors like [niobium diselenide], and how one can do wafer-scale stacking of these films,” added Wang.Wang believes that this research has shown 2D hBN to be a good insulator candidate for superconducting qubits. He says that the groundwork the MIT team has done will serve as a road map for using other hybrid 2D materials to build superconducting circuits.

It's FOSS

Jan 24, 2026

loss32 is a Linux Distro Nobody Asked For (But Still Has Its Usage)

Jan 24, 2026

Good News for Mobile App Developers: Skip Is Now Open Source

Jan 24, 2026

I (Briefly) Tried Proton Lumo's New AI Workspaces

Jan 23, 2026

cURL Gets Rid of Its Bug Bounty Program Over AI Slop Overrun

Jan 23, 2026

Not An April Fool's Joke, You Can Run Linux and Windows on This Android Smartphone

Jan 22, 2026

This Could be the Best Graphics Editor for Linux Users (Yes, it is Open Source)

Jan 22, 2026

FreeBSD is a No-Go for KDE's Plasma Login Manager

Jan 22, 2026

Espressif Launches "Radio Co-Processor" ESP32-E22

Jan 22, 2026

Open Source Project LLVM Says Yes to AI-Generated Code, But Not Without Conditions

Jan 22, 2026

FOSS Weekly #26.04: Mint 22.3 Released, Stallman is Back, Ubuntu's Snap Store Under Attack and More

Jan 21, 2026

I Ran the Famed Affinity Designer on Linux. Here's How It Went

Jan 21, 2026

Elecrow CrowPanel Rotary Display Review: A Smart Knob for DIY Enthusiasts

Jan 21, 2026

Richard Stallman to Speak on Software Freedom and AI at Georgia Tech

Jan 20, 2026

Ubuntu's Snap Store is Under Siege from Scammers, and The Gates are Open

Jan 19, 2026

Photoshop Installer Runs on Linux Despite Adobe Not Giving a Damn For Years

Hackaday

Jan 25, 2026

Ancient Egyptian Flatness

Jan 24, 2026

Wireless MIDI Controller Has Lots Of Knobs

<div><img width="800" height="450" src="https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png?w=800" class="attachment-large size-large wp-post-image" alt="" style="margin: 0 auto; margin-bottom: 15px;" decoding="async" loading="lazy" srcset="https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png 1280w, https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png?resize=250,141 250w, https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png?resize=400,225 400w, https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png?resize=800,450 800w" sizes="auto, (max-width: 800px) 100vw, 800px" data-attachment-id="905387" data-permalink="https://hackaday.com/2026/01/24/wireless-midi-controller-has-lots-of-knobs/twisty-2-an-open-source-bluetooth-midi-controller-using-the-raspberry-pi-pico-w-9-16-screenshot/" data-orig-file="https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png" data-orig-size="1280,720" data-comments-opened="1" data-image-meta="{"aperture":"0","credit":"","camera":"","caption":"","created_timestamp":"0","copyright":"","focal_length":"0","iso":"0","shutter_speed":"0","title":"","orientation":"0"}" data-image-title="Twisty 2 &#8211; an Open Source Bluetooth MIDI controller using the Raspberry Pi Pico W 9-16 screenshot" data-image-description="" data-image-caption="" data-medium-file="https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png?w=400" data-large-file="https://hackaday.com/wp-content/uploads/2026/01/Twisty-2-an-Open-Source-Bluetooth-MIDI-controller-using-the-Raspberry-Pi-Pico-W-9-16-screenshot.png?w=800" /></div>We live in a golden age for MIDI controllers. [rheslip]’s contribution to the milieu is a twisty take on the format, in that it’s covered in an array of knobs. <a href="https://hackaday.com/2026/01/24/wireless-midi-controller-has-lots-of-knobs/" class="read-more">…read more</a>

Jan 24, 2026

Reviewing Tweezers for Microsoldering and SMD Work

<div><img width="800" height="450" src="https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg?w=800" class="attachment-large size-large wp-post-image" alt="[nanofix] and his assortment of tweezers" style="margin: 0 auto; margin-bottom: 15px;" decoding="async" loading="lazy" srcset="https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg 1824w, https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg?resize=250,141 250w, https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg?resize=400,225 400w, https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg?resize=800,450 800w, https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg?resize=1536,864 1536w" sizes="auto, (max-width: 800px) 100vw, 800px" data-attachment-id="905230" data-permalink="https://hackaday.com/2026/01/24/reviewing-tweezers-for-microsoldering-and-smd-work/nanofix-tweezers-banner/" data-orig-file="https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg" data-orig-size="1824,1026" data-comments-opened="1" data-image-meta="{"aperture":"0","credit":"","camera":"","caption":"","created_timestamp":"0","copyright":"","focal_length":"0","iso":"0","shutter_speed":"0","title":"","orientation":"0"}" data-image-title="nanofix-tweezers-banner" data-image-description="" data-image-caption="" data-medium-file="https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg?w=400" data-large-file="https://hackaday.com/wp-content/uploads/2026/01/nanofix-tweezers-banner.jpg?w=800" /></div>Attributed to Picasso was the notion that when art critics get together they talk about content, style, trend, and meaning; but that when painters get together they talk about where <a href="https://hackaday.com/2026/01/24/reviewing-tweezers-for-microsoldering-and-smd-work/" class="read-more">…read more</a>

Jan 24, 2026

Crazy Old Machines

Jan 24, 2026

Isolated AC/DC Power Supply and Testing Station for 230 V Devices

<div><img width="800" height="486" src="https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png?w=800" class="attachment-large size-large wp-post-image" alt="" style="margin: 0 auto; margin-bottom: 15px;" decoding="async" loading="lazy" srcset="https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png 1360w, https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png?resize=250,152 250w, https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png?resize=400,243 400w, https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png?resize=800,486 800w" sizes="auto, (max-width: 800px) 100vw, 800px" data-attachment-id="905174" data-permalink="https://hackaday.com/2026/01/24/isolated-ac-dc-power-supply-and-testing-station-for-230-v-devices/variable_ac-dc_power_source_slomkowski/" data-orig-file="https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png" data-orig-size="1360,827" data-comments-opened="1" data-image-meta="{"aperture":"0","credit":"","camera":"","caption":"","created_timestamp":"0","copyright":"","focal_length":"0","iso":"0","shutter_speed":"0","title":"","orientation":"0"}" data-image-title="variable_ac-dc_power_source_slomkowski" data-image-description="" data-image-caption="" data-medium-file="https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png?w=400" data-large-file="https://hackaday.com/wp-content/uploads/2026/01/variable_ac-dc_power_source_slomkowski.png?w=800" /></div>When you’re testing or debugging some mains-powered gear, plugging it directly into the outlet can often be an exciting proposition. If such excitement is not really your thing, you can <a href="https://hackaday.com/2026/01/24/isolated-ac-dc-power-supply-and-testing-station-for-230-v-devices/" class="read-more">…read more</a>

Jan 24, 2026

Environmental Monitoring on the Cheap

Jan 24, 2026

Beer Keg Plumbing for a Liquid Rocket

<div><img width="800" height="450" src="https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png?w=800" class="attachment-large size-large wp-post-image" alt="" style="margin: 0 auto; margin-bottom: 15px;" decoding="async" loading="lazy" srcset="https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png 1920w, https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png?resize=250,141 250w, https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png?resize=400,225 400w, https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png?resize=800,450 800w, https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png?resize=1536,864 1536w" sizes="auto, (max-width: 800px) 100vw, 800px" data-attachment-id="904990" data-permalink="https://hackaday.com/2026/01/23/beer-keg-plumbing-for-a-liquid-rocket/plumbing-a-liquid-rocket-made-of-beer-kegs-kegrocket-ep-5-13-30-screenshot/" data-orig-file="https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png" data-orig-size="1920,1080" data-comments-opened="1" data-image-meta="{"aperture":"0","credit":"","camera":"","caption":"","created_timestamp":"0","copyright":"","focal_length":"0","iso":"0","shutter_speed":"0","title":"","orientation":"0"}" data-image-title="Plumbing a Liquid Rocket made of BEER KEGS &#8211; KegRocket Ep. 5 13-30 screenshot" data-image-description="" data-image-caption="" data-medium-file="https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png?w=400" data-large-file="https://hackaday.com/wp-content/uploads/2026/01/Plumbing-a-Liquid-Rocket-made-of-BEER-KEGS-KegRocket-Ep.-5-13-30-screenshot.png?w=800" /></div>When you think of a high performance liquid rocket, what do you think of? Beer kegs? No? Well, when [Ryan] from the YouTube channel “Project KegRocket” saw a beer keg, <a href="https://hackaday.com/2026/01/23/beer-keg-plumbing-for-a-liquid-rocket/" class="read-more">…read more</a>