In a nutshell: The European Commission is developing a new strategy to enhance internal security across the entire European Union. The plan introduces a cultural shift in how internal security is perceived, but some aspects of the proposal could potentially weaken digital security instead. The recently announced ProtectEU plan aims to safeguard Europe from internal threats, European Commission President Ursula von der Leyen stated, emphasizing that security is a fundamental pillar of open societies and thriving economies. Among the most pressing threats identified by the European Commission are terrorism, organized crime, cybercrime, and attacks on critical infrastructure.The EC is positioning ProtectEU as a broad initiative in response to an increasingly complex geopolitical landscape. With hybrid threats and state-sponsored cyberattacks wreaking havoc online, Europe urgently needs a significant overhaul of its internal security strategy. While ProtectEU does not yet include a detailed proposal, the official announcement outlines several key objectives as a foundation for the plan.ProtectEU advocates for a new security governance framework, aiming to transform the European Commission's approach to internal security. The initiative calls on EU member states to enhance intelligence-sharing practices through a newly proposed platform, the Single Intelligence Analysis Capacity. Additionally, the Commission envisions expanding Europol's role, effectively transforming it into a European equivalent of the FBI, with enhanced operational capabilities.The most intriguing and concerning aspect of ProtectEU from a technological standpoint is its focus on Europol's expanded enforcement tools, which are intended to address increasingly complex criminal investigations involving digital information. The plan includes provisions for "lawful access to data," outlining a technology roadmap that targets encryption and evaluates the EU's current data retention rules.Granting Europol the ability to access encrypted data can only mean one thing: Brussels is proposing some form of government-mandated backdoor for communication platforms protected by end-to-end encryption. These systems are notoriously difficult to breach, making them a challenge for law enforcement. However, security researchers have repeatedly demonstrated that implementing so-called "lawful" backdoors is inherently flawed as such vulnerabilities would inevitably be discovered, accessed, and exploited by cybercriminals and black-hat hackers. // Related StoriesIt remains to be seen how the EU plans to move forward with its encryption-breaking ambitions, though expectations for a practical or effective outcome are low. Meanwhile, the ProtectEU initiative also includes proposals to strengthen digital infrastructure, tighten regulations against organized crime networks, and expand international cooperation, particularly with Latin America and the Mediterranean region.Notably, the European Commission makes no mention of new partnership initiatives with the United States.

A little bit more to share: When Nintendo unveiled the Switch 2, it was vague about the chip powering the device, only mentioning that it was a custom design from Nvidia. It didn't list any specs besides general information - up to 120fps on device, 4K at 60fps, VRR, etc. Thankfully, Nvidia has shed a little more light on the subject. Nvidia has posted a blog highlighting the features offered by their hardware powering the Switch 2 beyond the information Nintendo provided. Among them is ray tracing support.While ray tracing is also available on other consoles, it took a while for developers to start actually using it. Spider-Man: Miles Morales and Cyberpunk 2077 were a couple of the first titles to implement RT in the PlayStation 5 and Xbox Series X. The Shortcut notes that as of January, the list of PS5 titles with RT support has grown to 58. However, some only offer ray tracing on the PS5 Pro version.Another bragging right Nintendo can claim is DLSS support, now a mainstay for desktop gaming. Team Red's platform-agnostic alternative, FidelityFX Super Resolution (FSR), has recently trickled into gaming laptops, consoles, and handheld PCs. Both technologies allow systems to increase image quality while maintaining performance, but Switch 2 is the first gaming handheld to offer Nvidia's arguably superior DLSS. Thanks to its Tensor Cores, Nvidia claims Nintendo's hybrid has 10 times the graphical performance of the original Switch.Elden Ring hits Switch 2 later this year, which would be a good testing ground for its DLSS and RT features.All of this looks great on paper, but here's the rub. While having DLSS and ray tracing on the Switch 2 is a plus, it's up to developers to implement it. We aren't just going to see games leveraging these features immediately. It took some time for studios to prioritize it in PS5 and XBSX games. Rockstar didn't add it to GTA 5 until 2022 two years after the consoles debuted. It will likely take even longer before developers implement them on a device with a comparatively smaller screen.Even considering the boost in resolution the Switch 2 offers while docked to a TV, there is a hard push to market the device as a handheld gaming unit. Its 8-inch display at 120fps in 1080p is more than adequate for that use case. So, developers probably won't rush to implement DLSS and RT in patches or new games for a while. Nvidia didn't mention any games in development that would implement either technology. // Related StoriesThe more pressing problems with the Switch 2 are its price point and that of its games. At $450 for sub-console performance, many Switch owners are taking a hard pass on the upgraded model. Nintendo's new $70 to $80 target for new and previously released games has also raised many fans' hackles. While the device is clearly a capable handheld, it is not on par with home consoles except in its price.

The big picture: Google's latest mid-range smartphone, the Pixel 9a, comes with a controversial new battery management feature that gradually reduces the battery's maximum charge capacity over time, whether you like it or not. While this might sound concerning at first, it could ultimately help extend the battery's overall lifespan. The feature, called "Battery Health Assistance," is essentially a software mechanism that automatically lowers the maximum charge level on your Pixel 9a after every 200 charge cycles. These reductions occur incrementally until the device reaches 1,000 cycles in total.At first glance, having your phone's OS unilaterally limit your battery capacity might seem alarming. However, Google insists that the feature is designed for the greater good.All lithium-ion batteries degrade over time. While your Pixel may start with 100 percent battery capacity when brand new, repeated charge cycles gradually reduce its ability to hold a full charge. Google's approach aims to slow down this natural wear and tear, potentially extending your battery's long-term health.In other words, your battery's capacity will decline over time anyway Google just wants to manage that process. By gradually reducing the maximum charge, the company claims Battery Health Assistance can help extend your battery's lifespan and slow down degradation.Ultimately, this approach should translate to longer usable battery life later into your phone's lifecycle. // Related StoriesGoogle does warn that you may notice "small decreases in your battery's runtime" as the battery ages. Additionally, the feature will adjust charging speeds based on the reduced capacity, which could lead to minor changes in charging performance. The full explanation is available in Google's official support article.Battery Health Assistance is mandatory on the Pixel 9a, and there's no option to disable it. It's also likely to be included in future Pixel models, though it's unclear if it will roll out to older devices via a software update.For context, Google previously stated that the Pixel 8a and newer models are designed to retain up to 80 percent capacity after approximately 1,000 charge cycles. In contrast, older Pixels (from the 3a to the 8 Pro) are rated for 80 percent capacity at 800 cycles. This new battery tech appears aimed at pushing sustainable battery life even further.Beyond the Pixel 9a's battery management feature, Android 16 is also expected to introduce deeper battery insights. The update should include a Battery Health menu displaying your phone's estimated remaining capacity percentage and detailed charge cycle statistics.

Editor's take: AI bots have recently become the scourge of websites dealing with written content or other media types. From Wikipedia to the humble personal blog, no one is safe from the network sledgehammer wielded by OpenAI and other tech giants in search of fresh content to feed their AI models. The Wikimedia Foundation, the nonprofit organization hosting Wikipedia and other widely popular websites, is raising concerns about AI scraper bots and their impact on the foundation's internet bandwidth. Demand for content hosted on Wikimedia servers has grown significantly since the beginning of 2024, with AI companies actively consuming an overwhelming amount of traffic to train their products.Wikimedia projects, which include some of the largest collections of knowledge and freely accessible media on the internet, are used by billions of people worldwide. Wikimedia Commons alone hosts 144 million images, videos, and other files shared under a public domain license, and it is especially suffering from the unregulated crawling activity of AI bots.The Wikimedia Foundation has experienced a 50 percent increase in bandwidth used for multimedia downloads since January 2024, with traffic predominantly coming from bots. Automated programs are scraping the Wikimedia Commons image catalog to feed the content to AI models, the foundation states, and the infrastructure isn't built to endure this type of parasitic internet traffic.Wikimedia's team had clear evidence of the effects of AI scraping in December 2024, when former US President Jimmy Carter passed away, and millions of viewers accessed his page on the English edition of Wikipedia. The 2.8 million people reading the president's bio and accomplishments were 'manageable,' the team said, but many users were also streaming the 1.5-hour-long video of Carter's 1980 debate with Ronald Reagan.As a result of the doubling of normal network traffic, a small number of Wikipedia's connection routes to the internet were congested for around an hour. Wikimedia's Site Reliability team was able to reroute traffic and restore access, but the network hiccup shouldn't have happened in the first place. // Related StoriesBy examining the bandwidth issue during a system migration, Wikimedia found that at least 65 percent of the most resource-intensive traffic came from bots, passing through the cache infrastructure and directly impacting Wikimedia's 'core' data center.The organization is working to address this new kind of network challenge, which is now affecting the entire internet, as AI and tech companies are actively scraping every ounce of human-made content they can find. "Delivering trustworthy content also means supporting a 'knowledge as a service' model, where we acknowledge that the whole internet draws on Wikimedia content," the organization said.Wikimedia is promoting a more responsible approach to infrastructure access through better coordination with AI developers. Dedicated APIs could ease the bandwidth burden, making identification and the fight against "bad actors" in the AI industry easier.

A hot potato: President Donald Trump has dropped what experts are calling a "nuclear bomb" on international trade. On Wednesday, he confirmed sweeping new tariffs of at least 10% on all imports into the United States with significantly higher duties targeting both allies like the European Union and economic rivals like China. Most trade experts believe the move will provoke further tit-for-tat tariffs, fueling a potentially devastating and deflationary global trade war with no clear winners. This aggressive escalation of Trump's trade wars could mean higher prices on virtually every product Americans purchase from overseas. Worse, analysts predict the tariffs could trigger recessions in multiple countries within the next year. "He just dropped a nuclear bomb on the global trading system," declared Ken Rogoff, former chief economist at the International Monetary Fund, in an interview with the BBC.As a reaction to the announcement, technology stocks plunged Thursday (figures above reflect early morning trading). Apple so far is leading the decline among Big Tech enterprises sinking nearly 9% its steepest drop since 2020 due to its reliance on overseas manufacturing.The Nasdaq Composite index dropped over 5%, on track for its worst single-day performance in more than five years, and is now down 14% year to date. Other major tech names also suffered: Meta and Amazon each fell more than 7%, while Nvidia and Tesla declined over 5%. Microsoft and Alphabet each slipped around 2%.Even though certain categories like computer chips, copper, and medicine are excluded from the new tariffs a special tax on semiconductors is expected to come at a later date chipmakers were hit hard as well. Marvell, Arm, and Micron Technology all plunged more than 8%. Broadcom and Lam Research dropped 6%, and AMD slid over 4%. PC makers Dell and HP suffered the most, with both stocks crashing more than 16%.Under the plan, a baseline 10% tariff will be imposed on all imports starting April 5. However, goods from roughly 60 countries labeled by the White House as the "worst offenders" on trade will face steeper penalties just a few days later, on April 9.These countries will be hit with what Trump described as "reciprocal" tariff rates, matching the higher duties and taxes those nations currently impose on U.S. exports. For example, the EU's 20% tariffs will be met with a new 20% U.S. tax on European goods. China, whose exports were already heavily taxed under Trump's previous trade actions, now faces a staggering total duty rate of 54%. // Related StoriesWhile neighboring Canada and Mexico avoided Trump's wrath this time, nearly every other U.S. trade partner was affected including long-time allies such as the UK (10% tariff) and Japan (24%). Some of the highest tariffs are aimed at countries like Vietnam (46%) and Cambodia (49%), which have experienced investment booms in recent years as companies shifted manufacturing from China to avoid earlier tariffs.However, many questions remain unanswered, including whether these new tariffs will actually create more U.S. factory jobs, as Trump claims.It's also worth noting that Apple and other tech giants that have been diversifying away from China could now get hit with high tariffs regardless of which country they use for manufacturing products sold in the U.S.Economists had previously warned that tariffs would lead to higher prices on imports, including electronic products such as laptops, smartphones, monitors, computers, and TVs, most of which are primarily manufactured in China.Retailers like Amazon, which rely on ultra-cheap goods shipped tariff-free in small packages, were also hit hard. Trump has moved to reinstate plans to kill a long-standing exemption that allowed items valued under $800 to enter the U.S. duty-free. The change could take effect in May.Unsurprisingly, both China and the EU have threatened retaliation. However, Treasury Secretary Scott Bessent has urged against it, insisting the new tariffs represent the "high water mark" that other nations should simply accept.

The big picture: The AI boom is significantly changing the enterprise and data center market. Arm predicts that chips based on its designs will soon amount to around 50% of CPUs used in data centers. According to Arm Senior VP Mohamed Awad, the company's market share in this sector will grow by 15 percent over 2024. Chips based on the Arm architecture used to play a marginal role in the data center business. However, things are changing fast and Arm expects substantial financial rewards thanks to the company's key partners in the market.Arm struggled to compete against the x86 enterprise leadership for nearly 20 years due to significant hardware infrastructure changes and the associated costs. Reuters notes that Arm CPUs have increased in popularity in data centers thanks to their lower power consumption levels compared to x86 chips made by Intel and AMD.Compounding the problem are the growing energy requirements of AI processing. Cloud computing providers desperately seek alternatives, and software solutions are helpful. "We've gotten to the point where software is actually being developed for Arm first and foremost," Awad said.Nvidia is another major factor in this significant market evolution, as the GPU maker heavily relies on Arm's chip architecture and IP. The Grace CPU is a custom-built solution that fully exploits the computing capabilities of the Blackwell framework.Nvidia Grace's features and benefits include better energy efficiency, ECC LPDDR5X memory for improved reliability, and a "tightly coupled" CPU and GPU system architecture. // Related StoriesAwad explained that chips designed for data centers must achieve higher complexity and efficiency to compete since Arm's business model is primarily licensing-based. The company typically receives much greater aggregate royalty rates on higher-level designs than microchips for simpler devices and low-level consumer electronics.The data center market is one of Arm's key investments for the future. The Japanese-British chip designer had begun a substantial effort to differentiate its business, with management considering direct chip manufacturing as one solution. Arm is also in an antitrust legal battle with former business partner Qualcomm, though a victory is anything but guaranteed.

The big picture: An April Fools' prank is all in good fun unless it is taken too seriously. A Welsh journalist discovered that internet jokes in the AI age aren't as funny. His harmless little gag demonstrates how AI picks up on new false information and proliferates it as factual. Every April 1st, Ben Black loves pulling off a prank on the readers of his community news site Cwmbran Life. The 48-year-old journalist has been spinning up April Fools' Day hoaxes over the years that range from the hilarious to the totally absurd.In 2020, he made up a story about Cwmbran setting a Guinness World Record for having the most roundabouts per square km, which took on a life of its own. A day after it went live, he had to update the article to clarify that it was an April Fools' joke. However, that didn't stop a more prominent publication from picking up the story and publishing it as "news." Even worse, Black requested the site to take the story down. It refused, but that was human error.Fast forward to 2025, and Ben was shocked to find that Google's AI Overview had used his fake story as the basis for its summary and presented it as fact. While it's not dangerous misinformation, the incident is a prime example of how easily fake news can spread, even from trusted sources."It's really scary that someone googling 'roads in Wales' could come across my made-up story and believe it," Black told the BBC.Black also feels the incident exposes how AI represents a growing threat to small, independent publishers. While major outlets strike lucrative deals to collaborate with AI companies, he says smaller ones like his get left out in the cold as their original content gets mined and repurposed without permission.Of course, the concept of satirical news on the internet isn't new. Outlets like The Onion and The Babylon Bee have been around for ages. Large language models are likely "aware" of these well-known examples and perhaps others and treat them differently than real news. However, considering Cwmbran Life is an actual news site, the AI couldn't tell the difference and took the prank seriously. // Related StoriesSadly, the mess convinced Black to hang up his April Fools' Day tradition for good. Maybe developers should program LLMs to seek out further context before reporting "facts," especially on April 1st.Image credit: BBC

Forward-looking: Engineers at the University of California, Berkeley, have unveiled a flying robot that mimics the agility and precision of a bumblebee. Weighing just 21 milligrams and measuring less than one centimeter in diameter, it is the world's smallest wireless robot capable of controlled flight. "Bees exhibit remarkable aeronautical abilities, such as navigation, hovering, and pollination, that artificial flying robots of similar scale fail to do," explained Liwei Lin, distinguished professor of mechanical engineering at UC Berkeley and senior author of the study. "This flying robot can be wirelessly controlled to approach and hit a designated target, mimicking the mechanism of pollination as a bee collects nectar and flies away." The research was published in Science Advances.Designing a small-scale flying robot posed significant challenges. Traditional flying robots rely on onboard power sources like batteries and electronic flight control systems components that are difficult to miniaturize without adding excessive weight. To overcome this hurdle, Lin and his team used an external magnetic field to both power the robot and control its movements.The robot's design is deceptively simple yet highly effective. Shaped like a tiny propeller, it features two small magnets that respond to an external magnetic field. As the field alternates, the magnets are attracted and repelled, causing the propeller to spin. This spinning motion generates enough lift for the robot to take flight. By adjusting the field's strength, researchers can precisely control its flight path.At just one centimeter in diameter, this device is nearly three times smaller than its closest competitor, a 2.8-centimeter-diameter flying robot. Its compact size enables applications in environments where larger robots cannot operate."Tiny flying robots are useful for exploring small cavities and other complicated environments," said Fanping Sui, co-first author of the study and a recent Ph.D. graduate from UC Berkeley. Potential uses include artificial pollination and inspecting confined spaces, such as inside pipes. // Related StoriesHowever, the robot currently has limitations. It operates in passive flight mode, meaning it lacks onboard sensors to monitor its position or trajectory in real time. As a result, sudden environmental changes such as strong gusts of wind could disrupt its course.Wei Yue, co-first author and a graduate student in Lin's lab, noted that future iterations aim to incorporate active control systems, enabling real-time adjustments to position and attitude.Another challenge is the robot's reliance on a strong magnetic field generated by an electromagnetic coil for operation, although researchers believe that further miniaturization could address this issue. Reducing the robot's size to less than one millimeter could allow it to be powered by weaker magnetic fields, such as those from radio waves.The bumblebee-inspired innovation is not the only achievement of Lin's team in bio-inspired robotics. They have also developed a cockroach-like robot (above) capable of withstanding extreme forces, such as being stepped on by a human.Meanwhile, Yue is leading efforts to create "swarming" robots that can collaborate like ants to perform complex tasks beyond the capabilities of individual robots."I'm working with 5-millimeter-scale robots that can crawl, roll, and spin," Yue explained. "They can also work together to form chains and arrays or tackle harder tasks."These swarming robots hold promise for applications such as minimally invasive surgery, where multiple units could be injected into the body to form stents or remove clots collaboratively.Kamyar Behrouzi, Yuan Gao, and Mark Mueller from UC Berkeley co-authored the study, which was supported by the Berkeley Sensor and Actuator Center.

Facepalm: When ChatGPT isn't being used to generate those viral Studio Ghibli-inspired images flooding your feeds, it can be tapped to fake receipts instead. The AI chatbot is finally decent enough at comprehending and generating text within images that crafty users realized it could produce fake receipts. These could then easily pass as legit proof of purchase to an unsuspecting set of eyes. Until recently, AI-generated images had one glaring weakness garbled or nonsensical text. Many models have managed to fix problems like messed-up hands to the point where they're not even an issue anymore. But text has remained a bit of an Achilles' heel. That's changing fast, though.For instance, following the latest update to ChatGPT's image generation on its 4o model, users can spit out receipts with near-perfect formatting, itemized lists, and even realistic-looking logos. Of course, they're still not completely flawless and can look overly crisp, have text with odd spacing, and even have math errors.However, receipts especially for smaller amounts aren't often scrutinized too heavily. A few extra tweaks in Photoshop, like adding stains, could easily fool an employer. The payoff is obvious to scammers. All they'd need to do is spend just a few minutes using ChatGPT and Photoshop, and even if the trick only works occasionally, it's still a low-effort, high-reward scheme.Right now, the most effective method might be to feed an existing receipt to ChatGPT and modify details like prices or dates. But as the tech improves, entirely fabricated receipts could become indistinguishable from real ones. // Related StoriesYou'd think that all this would ring alarm bells at OpenAI HQ, but while the company has acknowledged the risk, it isn't hitting the panic button just yet. A company spokesperson explained to TechRadar that it tracks image generation both on and off its platform, uses internal tools to verify AI-created content, and takes action when usage policy violations occur.Interestingly, OpenAI also sees an upside to this capability. Spokesperson Taya Christianson suggested to TechCrunch that fake receipts could be used for financial literacy education by teaching people how to track expenses or spot fraud in a controlled setting.Still, one has to wonder if this capability results in a net positive for society.

In a nutshell: It should go without saying that buying a very cheap, obviously counterfeit Android handset is a bad idea. Not only will you likely run into a slew of technical problems, but it could also contain pre-installed malware that infects virtually every process carried out by the handset. Kaspersky researchers have discovered a new strain of the Triada Trojan preinstalled on thousands of new very cheap Android devices counterfeit versions of popular models. The modular remote access trojan (RAT), first identified in 2016, can steal sensitive data and crypto while performing other malicious actions.The malware is embedded in an infected phone's firmware, operating without detection and granting attackers full control over a device.The list of malicious actions the new strain of Triada can perform is an extensive one. It is able to send and delete messages in WhatsApp, Telegram, and other messaging apps to impersonate users, hijack social media accounts, replace crypto wallet addresses in apps to steal the virtual currency, and track browsing activity and even swap links.Triada can also monitor, intercept, send, and delete SMS messages, send premium SMS texts to receive paid services, download and run other programs on an infected device, and block network connections to interfere with the operation of anti-fraud systems.Dmitry Kalinin, an expert on cybersecurity at Kaspersky Lab, said the new version of Triada is likely being installed in the counterfeit phones during one of the early stages of the supply chain, so the stores selling the devices may not realize anything is amiss. There's always the chance that some vendors knew about the malware and were benefiting financially by spreading it, of course.Kalinin adds that the authors of Triada have already used the malware to transfer about $270,000 in different cryptocurrencies from victims' crypto wallets to their own, though the actual amount may be higher as the attackers also stole Monero, a cryptocurrency that cannot be traced. // Related StoriesOver 2,600 people across multiple countries encountered the new version of Triada between March 13 and 27, 2025, with most of the cases reported in Russia.Kaspersky says its mobile anti-virus software can detect Triada, though the company's products are banned in the US over its ties to Russia.

In a nutshell: A Minecraft Movie, the live-action-with-CGI adaptation of the best-selling video game of all time, is released this Friday. The pretty underwhelming trailers led to plenty of people calling this the next Borderlands, but according to early reviews and first impressions, it's actually okay. Movie adaptations of video games have come a long way since Bob Hoskins and John Leguizamo horrified the world with Super Mario Bros. in 1993. But while Hollywood does pump out the likes of Sonic the Hedgehog these days, there are painfully bad adaptations like Borderlands, which has a 10% Rotten Tomatoes score, still being made.The trailers for Minecraft, starring Jack Black and Jason Momoa, led plenty to believe this would be another poor cash-in with no reverence for the source material. But apparently not.Reviews ahead of Friday's release range from it being "good, actually" and "surprisingly specific and funny," to "baseline entertaining" and not actively boring. Even often-harsh The Guardian calls it rollicking if exhausting fun, awarding the movie three out of five stars.One of the most critical reviews is from the Hollywood Reporter, which calls it an unimaginative film that fails to spark the imagination.Right now, The Minecraft Movie has a 53% critics rating on Rotten Tomatoes from 51 reviews, while its IMDB score is 6.2. // Related StoriesAs is so often the case with movies of this ilk, a lot of the general public seem to like the Minecraft movie more than the critics.Several viewers mention that the film could join Napoleon Dynamite and Nacho Libre, other movies where director Jared Hess was behind the camera, in becoming a cult classic as it features quirkiness, awkward humor, and weirdness in abundance.It's also been pointed out several times that adults aren't really the primary target demographic for A Minecraft Movie. And while plenty of over 18s say they enjoyed it or at least didn't hate it expect most kids to love it.Being rated "not as bad as you expected" and "okay, sometimes good" certainly doesn't mean Minecraft will lose money. The top ten highest-grossing films based on video games include Rampage ($428 million), Uncharted ($407 million), and Prince of Persia: The Sands of Time ($336 million), all of which received similar mixed reviews upon their release.The Minecraft Movie opens in the US on April 4.

In a nutshell: Laptops featuring Nvidia's high-end RTX 50 series graphics cards have started rolling out, and early reviews suggest the company's latest flagship offers a somewhat lukewarm uplift over its predecessor. Furthermore, a new series of benchmarks indicates that moving down one rank might provide very similar performance while saving customers hundreds of dollars. Notebookcheck recently tested a mobile Nvidia RTX 5090 and 5080 paired with the same CPU, providing a nearly like-for-like comparison. Although some laptops featuring the 5080 are up to $1,000 cheaper than 5090-equipped notebooks, the former is often less than 10 percent slower.Using XMG Neo 16 A25 engineering samples with AMD Ryzen 9 9955HX processors, the RTX 5080 stuck closely to its larger sibling in four 3DMark tests and several high-end games. The biggest gap appeared in Steel Nomad, where the 5090 scored 6,156 18 percent above the 5080. Its advantage shrank to 15 percent in Port Royal at 1440p and under 10 percent in other benchmarks.Unsurprisingly, the RTX 5090 pulled the furthest ahead in gaming at 4K, while the difference between the GPUs diminishes considerably at 1080p in most titles. For example, the 5080 and 5090 perform identically at 1080p in Final Fantasy XV, suggesting that the burden may have mostly shifted to the CPU. In most other 1080p benchmarks, the 5090's lead hovers around five percent. In 4K, it extends to roughly 15 percent in games like Cyberpunk 2077 and Baldur's Gate 3.However, Monster Hunter Wilds, Alan Wake 2, and Assassin's Creed Shadows challenged both GPUs. In native 4K with ray tracing, even the laptop 5090 couldn't reach 50 fps in Monster Hunter, and the 5080 fell into the low 40s. Both cards collapsed to below 20 fps in Alan Wake 2 with ray tracing at 4K, and the 5090 maintained a 13 to 15 percent advantage at multiple resolutions in Assassin's Creed, including 1080p.Most early reviews suggest that the mobile RTX 5090 is only worthwhile for customers upgrading from the RTX 30 series or earlier, as it only gains around 10 fps over the 4090 in most games. The results from the 5080 likely give it a far superior cost-per-frame value. // Related StoriesLaptops featuring the mid-range RTX 5070 and 5070 Ti are expected to appear in the coming weeks, and Nvidia is set to launch the desktop 5060 and 5060 Ti soon.

What just happened? AMD has responded to reports of 108 Ryzen 7 9800X3D CPUs unexpectedly dying in users' machines, with most of the incidents happening on Asrock motherboards. Team Red blames memory compatibility issues for the processors "failing to complete POST," even though the affected chips reportedly did pass POST and worked for anything from half an hour to several months before dying, with many cases showing physical damage. This week saw a Redditor compile a list of 108 Reddit posts involving a Ryzen 7 9800X3D that died, along with 12 other Ryzen 9000-series chips that suffered the same fate. 98 of the failures, or 82%, occurred on Asrock boards. It's stated that the processors passed POST and worked for varying amounts of time before dying with no signs of failure.That many instances doesn't represent a huge number compared to the thousands of Ryzen 7 9800X3D CPUs sold each week the processor is incredibly popular with gamers, and for good reason. But it's not an insignificant figure, either, so AMD gave an official response to some outlets.Also read: AMD Ryzen 7 9800X3D Review: The New Gaming CPU King"We are aware of a limited number of user reports involving Asrock AM5 motherboards failing to complete POST. Following a joint investigation, AMD and Asrock identified a memory compatibility issue present in earlier BIOS versions, which has been rectified in the latest BIOS," the company wrote, repeating previous statements made by Asrock.AMD added that failure to POST can be caused by several factors and does not necessarily indicate a non-functional CPU. It recommends that anyone experiencing problems start by updating their BIOS to the latest version for their motherboard. If issues persist, the next step is to contact AMD's customer support team. // Related StoriesAsrock responded to the reports last week. The company said that it had inspected a motherboard from a system where the CPU showed burn damage. It claimed there was no burn damage around the mobo or VRM area. After "cleaning and removing debris" from the CPU socket, the motherboard booted up successfully with the original BIOS and passed long-term stress tests.AMD is saying that a memory compatibility issue prevented the CPUs from POSTing, stopping them from working entirely as one would expect. But the Reddit post states that these problematic chips are passing POST and working for a while before failing, sometimes with electrical or burn damage, which doesn't sound like a symptom of the issue AMD is blaming.One can only conclude that AMD's statement is referring only to the cases where the 9800X3Ds didn't pass POST, and not those that suffered permanent physical damage. Nevertheless, a lot of user response to the explanation has been negative. It's possible that the company will follow-up with another statement at a later date.

Why it matters: As ByteDance's April 5 deadline to divest TikTok or face a ban approaches, last-minute players have entered negotiations to potentially take over the U.S. operations of the popular social network. The Trump administration, which postponed the ban in January, is reportedly exploring several options to bring some or all of TikTok under American ownership. Sources recently told The New York Times that Amazon made an 11th-hour offer to purchase all of TikTok, which faces a ban this Saturday April 5 if its Chinese owner, ByteDance, fails to divest the app to a U.S. buyer. Other new bidders include the mobile app platform AppLovin and a company co-founded by OnlyFans founder Tim Stokely.Amazon's offer reportedly came in the form of a letter addressed to Vice President JD Vance and Commerce Secretary Howard Lutnick, but officials involved in the negotiations are not seriously considering it.Meanwhile, OnlyFans Stokely's new social media company, Zoop, is working with the cryptocurrency foundation HBAR on a bid that aims to alter TikTok's revenue-sharing model.Congress passed a law last year banning the social media platform which is used by roughly half of all Americans unless a U.S.-based buyer emerges. Many worry that China's government could compel ByteDance to hand over American user data or manipulate TikTok's recommendation algorithm to spread propaganda. However, a potential deal could still leave the algorithm under Chinese control.At least one proposal under consideration by the Trump administration would have ByteDance lease the algorithm to U.S. companies while retaining a minority stake. Potential investors include tech giant Oracle and private equity firm Blackstone. It remains unclear whether Beijing or U.S. lawmakers would approve such a plan. // Related StoriesPresident Trump initially attempted to force ByteDance to divest TikTok by threatening a ban during his first term, but later reversed course. After President Biden signed last year's law, Trump extended the legislation's original deadline to April 5 just days after taking office and recently stated that a deal is imminent.AI startup Perplexity has also proposed acquiring and overhauling TikTok. The company would rebuild the algorithm, introduce community notes, and open-source the recommendation system.Others who have reportedly expressed interest in the platform include YouTuber MrBeast, Reddit co-founder Alexis Ohanian, Los Angeles Dodgers owner Frank McCourt, tech entrepreneur Jesse Tinsley, and Wyoming businessman Reid Rasner.

Finally: This morning's Switch 2 exclusive Nintendo Direct finally answered all the burning questions about the device, including key specs, price, and release date. As expected, many of the rumors regarding the handheld were validated, but let's make it official. The Nintendo Switch 2 is significantly bigger than its older sibling, featuring a 7.9-inch capacitive touch display, adding to its heft compared to its 6.2-inch predecessor. The 1080p IPS LCD has 120Hz variable refresh rate technology and HDR support.Nintendo was tight-lipped about the Switch 2's processor. The only detail it has divulged is that it is a custom-built CPU/GPU from Nvidia. Obviously, it's a step or two up from the first-gen switch, as evidenced by its shortened battery life. Users can expect between two and 6.5 hours of playtime, depending on the game. By comparison, the Switch could go about 4.5 to nine hours on a charge.Nintendo bumped the Switch 2 to 256GB of internal memory for eight times the storage, which should delight game hoarders like myself. It still has a slot for expanding storage but has switched from microSD to microSD Express. Even though it now has more storage, Nintendo has not abandoned physical game cards. The new mini-cartridges are virtually identical to the first-gen's but are red instead of gray. Additionally, last week's Nintendo Direct unveiled "Virtual Game Cards" to share your digital purchases just like physical media.The Switch 2's JoyCons are significantly different as well. The controllers are a bit bigger, as necessitated by the system's larger screen. However, the Switch 2 and its Joy-Cons (now branded "Joy-Con 2") maintain the same 13.9mm thickness of the original unit. I'm on the fence about whether this is good or bad.On the one hand, keeping it slim minimizes weight, which is essential for a handheld device. On the other, its thin profile is one of my nitpicks about the first-generation Switch. I always found it uncomfortable to use the controllers while attached. I was constantly having to adjust my grip because of its slim design. So, I mainly used the Joy-Con grip attachment for a better feel.Nintendo also confirmed the rumors that Joy-Con 2 controllers attach magnetically to the screen. There were initial concerns that this was a flimsy way to connect them, but that doesn't look like much of a problem with the Joy-Con 2's deep inset sockets and powerful internal magnets. The controllers should remain securely in place until the release lever is activated, no matter how much a user fidgets while playing. // Related StoriesOther nifty features include mouse mode and the "C" button. Mouse mode is just as it sounds. Users can position a single JoyCon sideways on a flat surface like a mouse. It's ideal for puzzle games, golf, or other compatible titles. The C button, which also appears on the new Nintendo Pro Controller, launches a self-explanatory new feature called GameChat and pulls up its related options. Switch's lack of a proper chat function was a big complaint, so it's nice to see Nintendo listen to user feedback.The last notable change to the Switch 2 is its docking system. The first thing users will notice is that the dock is much thicker. This extra thickness accommodates a cooling fan. Docking blocks airflow, and users typically play games longer while docked. Additionally, the hybrid supports 4K HDR at 60fps when connected to a TV, which probably increases heat generation. So, a fan to vent excessive heat makes sense.Pre-orders for the Switch 2 open next Wednesday, April 9, for $450. Customers can also buy the console bundled with Mario Cart World for $500. Units ship to pre-order holders and retailers on June 5. Check out the Nintendo Direct breakdown below for more details, including the new Switch 2 Camera.

What just happened? The Ultimate Fighting Championship has signed a multi-year, multimillion-dollar partnership that'll leverage Meta's various technologies to help the mixed martial arts organization better connect with fans. The wide-ranging deal will involve Meta platforms and services including Facebook, Instagram, WhatsApp and Threads. The UFC will also have access to Meta Quest, Meta Glasses, and Meta AI, potentially paving the way for groundbreaking fan experiences. Speaking of Threads, Meta's answer to X will now be the official social media partner of the UFC so don't expect to see X messages featured on broadcasts any longer.Meta, meanwhile, will receive favorable brand placement during PPV and Fight Night events in the octagon and be featured during broadcasts. Fans attending live events will also be able to take part in "creative" experiences, we're told.The companies said additional elements of the partnership will be announced as the two sides work together on future innovations.The partnership isn't entirely surprising considering the relationship between UFC President Dana White and Meta boss Mark Zuckerberg. In 2022, Zuckerberg started training in mixed martial arts and it wasn't long before we started seeing him sitting cageside at UFC events. He even appeared in the corner of Alexander Volkanovski at UFC 298 last year.In mid-2023, a potential MMA bout between Zuckerberg and Tesla CEO Elon Musk was floated but nothing ever came of it. Given the current climate surrounding Musk, perhaps White should revisit the matchup? They'd sell a ton of pay per views. White joined Meta's board of directors in January. // Related StoriesThe UFC was founded in 1993, with the original idea being for a tournament featuring various martial arts disciplines to see which was truly supreme. The promotion struggled mightily for years as critics deemed it too violent. US Senator John McCain famously described it as "human cockfighting."White was brought on in the early 2000s and eventually righted the ship. In 2016, WME-IMG purchased UFC parent company Zuffa for $4.025 billion.Image credit: Andrius Petrucenia

Editor's take: I will never understand why some people are so utterly obsessed with life simulation games. The entire concept behind The Sims and similar series strikes me as atrocious and meaningless a complete waste of time reduced to a soulless, algorithmic approximation of dull bot management. Maybe it's the optional adult-themed antics. After all, porn can explain and sell almost anything. Inzoi is a new life simulation game being developed in South Korea, using Unreal Engine 5 to deliver realistic graphics and a "life-like" relationship simulation. Inzoi Studio is clearly aiming to bring The Sims 4 into the modern tech era, though early reviews suggest the game could also turn out to be a dud. In any case, a certain type of gamer is already drooling over Inzoi's potential for virtual nudity and other adult content.Korean publisher Krafton released the early access edition of Inzoi on March 28, and modders are already experimenting with the new interactive sim. The Inzoi Mod Enabler, which provides a quick and easy way to bypass restrictions and load custom content, has already been downloaded more than 43,000 times.The second most-downloaded mod is Remove Shower Blur, which replaces the game's blurry effects with a more transparent look, allowing "Zois" characters to be properly displayed in towels when they enter the bathroom or take a shower. While the RSB mod doesn't enable any explicit NSFW content (yet), it serves as a reminder of how quickly modders can transform games into a 3D adult playground.Thanks to the recently released Inzoi Mod Enabler, gamers can now install genuinely useful tweaks, such as skipping startup videos, disabling annoying visual effects, and more. Of course, adult content is inevitable, and a developer is already offering a paid mod via Patreon to enable nude female Zois.Nude mods like WickedWhims are one of the key reasons why The Sims remains a popular series to this day. Inzoi could follow in The Sims 4's footsteps, offering gaming voyeurs a new way to experience naked 3D bots. // Related StoriesInzoi is an ambitious project, and Unreal Engine 5 is certainly capable of rendering realistic characters in a domestic environment, doing all the weird things sim bots tend to do. However, the game could also turn out to be a complete waste of time, as Zois appear to be obsessed with becoming influencers.

Forward-looking: The NFL has announced it will use Sony's high-tech Hawk-Eye cameras instead of traditional chains to measure the line to gain. The new technology is expected to speed up the process of determining first downs, but it won't necessarily take human error out of the equation. The change will take effect in the 2025 26 season, beginning with the kickoff game on September 4 in Philadelphia. The NFL expects the technology to add "a new level of precision and speed" to the decision-making process, offering a faster and more advanced alternative to the long-standing method of walking chains onto the field to manually measure the ball's distance from the line to gain.Hawk-Eye technology, which uses six 8K cameras for optical tracking of the ball's position, will be deployed at all 30 NFL stadiums in the U.S., as well as at international venues hosting NFL games. The system will be operated from the NFL's Art McNally GameDay Central Officiating Center (AMGC) in New York and is integrated with the league's existing replay system.According to the NFL, the new measurement process will take only about 30 seconds potentially saving up to 40 seconds compared to the traditional chain-based method. However, the chain crew won't be retired just yet. They will remain on the sidelines as a backup, and will be called upon to do the measuring in traditional style in case something goes wrong with the technology.TV audiences and in-stadium spectators will be able to view virtual recreations of measurements in real time, adding a new visual dimension to the game. Still, questions remain about whether the high-tech system will lead to more accurate calls.Fans have noted that the most controversial line calls typically stem from the initial spotting of the ball by on-field officials, rather than errors in measuring the distance to the line to gain. One example is last season's AFC Championship Game between the Buffalo Bills and Kansas City Chiefs, when Bills quarterback Josh Allen was controversially ruled short of the line to gain by the on-field officials on a critical play. // Related StoriesThe call was reviewed, but unfortunately for the Bills, the position of the ball could not be determined by the cameras, leading to the confirmation of the on-field ruling. Hawk-Eye would not have been of use in this instance, as the positioning of the ball by the on-field referees was the controversial aspect, not the first down measurement.Hawk-Eye technology is already used by 23 of the world's 25 top sports leagues and international sporting events, including various World Cups. It plays a key role in sports such as soccer, tennis, cricket, and rugby, where it helps determine line calls. Proponents argue that the technology has matured significantly in recent years, offering a more accurate and fairer experience for athletes, officials, and fans alike.

Something to look forward to: Meta's love of smart glasses is set to continue with a new device codenamed Hypernova that sounds distinctly cyberpunky. The glasses will feature a tiny heads-up display integrated into one of the lenses, which partly explains why it will cost anywhere between $1,000 and $1,400. The Hypernova smart glasses will arrive later this year, according to Bloomberg's Mark Gurman, so we could see them unveiled at the Meta Connect conference scheduled to take place on September 17 and 18.The big selling point could be the miniature monocular panel in the right-hand lens. It's expected that this screen could show app notifications from a user's phone, let them access their photos, or even see directions from an app such as Google Maps.The glasses' functions will be controlled using hand gestures and capacitive touch on the side of the frame. Gurman writes that the screen is in the lower-right quadrant of the right lens and is best viewed when looking downward. When the glasses are turned on, a home screen will load up and show icons horizontally, much like the Meta Quest. Expect to interact with Meta's AI assistant while wearing them, too.Hypernova will also have a more advanced camera system than the Ray-Ban Meta smart glasses, able to capture higher-quality photos than the 12MP sensor found on that device, which costs $300 or around $380 for versions with transition lenses.Meta's next-gen glasses will have to offer something quite special for users to spend anywhere from $1,000 to $1,400 on a pair. The mini screen does sound intriguing, but it's something that Google tried with Google Glass in 2013. It used a mini-projector that beamed images onto a semi-transparent prism positioned above the user's right eye, displaying a 640 360 image. Google discontinued the consumer version of Google Glass in 2015, while the Enterprise Edition was discontinued in 2023.Hypernova is different from Orion, the in-development augmented reality glasses from Meta. Those feature eye-tracking, hand-tracking, voice controls, and even a neural interface although it reads signals from your wrist rather than your brain. It's believed that Hypernova will come with this same wristband controller, codenamed Ceres.Meta seems confident that Hypernova will be a success, as it is also working on a successor, Hypernova 2, that will feature two screens and could launch in 2027. // Related Stories

What just happened? Google is celebrating Gmail's 21st birthday by introducing new encryption features for enterprise users. The company's latest security system aims to simplify encryption options for businesses of all sizes while providing IT departments with the tools to maintain information integrity. Other updates from this week improved Google Chrome's encryption and allowed Google Fi subscribers to send encrypted RCS messages to iPhones. Starting today, IT departments participating in Google's beta can use Gmail to send end-to-end encrypted emails to recipients within the same organization. The functionality will expand to all Gmail inboxes in the coming weeks and to non-Gmail inboxes by the end of the year.The new functionality, which is activated in just a few clicks, aims to provide small organizations with email security that typically requires dedicated IT departments. It also circumvents the need for proprietary software and exchanging decryption keys.The animation above from Google's Workspace blog demonstrates the process. To send an encrypted email, simply draft a new message, click the lock icon on the right side of the recipient field, and select "Turn on" under "Additional encryption."Emails are encrypted locally on the sender's device and decrypted in the recipient's inbox. The user experience closely resembles using Gmail normally, and Google's servers can't access the encryption keys. Meanwhile, non-Gmail inboxes that receive Gmail-encrypted messages receive an invitation to view a restricted version of Gmail. The new system also supports S/MIME.IT departments can also tighten security by forcing all recipients, even Gmail inboxes outside an organization, to view messages in Gmail's restricted mode, which presents the content like a Google Drive document. Security managers can set messages to encrypt by default and set rules for automating clearance levels to add extra layers of security without significantly inconveniencing users. // Related StoriesGoogle also introduced new encryption rules for Chrome this week, which should close certain HTTPS protocol loopholes. The tightened standards might significantly improve overall internet security, since Chrome is easily the most popular browser engine.In other Google-related encryption news, Apple's recently released iOS 18.4 update allows iPhones to receive RCS text messages from Google Fi subscribers. Before Apple introduced iOS 18 last year, messages sent between iPhones and Android devices lacked end-to-end encryption and other advanced features. The update that introduced cross-platform RCS initially supported only AT&T, T-Mobile, and Verizon users, but this week's patch extends support to Google Fi.

In brief: Valve has just posted the results of March's Steam survey, and, as predicted, things have returned to normal after the weirdness we saw during the previous month. This means there's a new top GPU, operating system, and language, while AMD's CPU share has reached a record high. February was one of those months where the Steam survey changed drastically. One of the biggest changes was a new most-popular graphics card, as the RTX 4060 replaced the RTX 3060.These sorts of anomalies can occur for several reasons, including sampling bias, reporting bugs, and hardware driver updates. There's also the problem of shared machines at internet cafes in China inflating hardware specs.But the issue is often down to Valve's algorithms and data processing methods, which can directly impact the survey results, especially when they make back-end changes or adjust how they categorize, interpret, or weight the incoming data.But, as we saw after the last survey shake-up in October 2023, the status quo has now been restored. That includes the RTX 3060 taking back its crown from the RTX 4060, though the Ampere card is only 0.34% ahead of its Lovelace successor in terms of user share.Most popular GPUs among Steam survey participants in March 2025March also saw AMD get back to business in the CPU section. Team Red hit a record 36.19% user share in January before falling 5% as Intel rose by the same amount a month later. In March, AMD reached a new record high of 37.62%, up 6.5%, with Intel dropping by the same margin to 62.3%. // Related StoriesWindows 10 also saw an unlikely resurgence in February, retaking the top spot with a 10% rise in user share. The older OS is now down to 40% and Windows 11 is found on 55% of participants' machines.This is a trend we're seeing globally as Windows 10's October 2025 end-of-support date draws closer. Statcounter reports that while the OS remains the most popular on desktops, its share has declined since the start of the year as Windows 11 climbs rapidly. There is now just 10 percent points separating the two operating systems.Returning to the Steam survey, 16GB has replaced 32GB as the most common amount of system RAM among participants, and English took over from Chinese as the most popular language.

Why it matters: In all the focus on the big processors and the "AI race," we often overlook the far less glamorous business of actually building compute systems. We have explored different parts of that (and here), but there is an underlying assumption built into all these conversations the role of Taiwanese companies upon which the whole ecosystem rests. What will the shift to AI compute mean for the hundreds of companies that make up this critical ecosystem? We were reminded of this during our recent visit to Mobile World Congress when we had the chance to visit the Taiwan ITRI Booth.ITRI is a trade organization representing their electronics companies. The booth had a dozen vendors offering various pieces of computing gear. This includes companies like iBase and Nexcom, which make specialty server gear (security and edge), Asrock, which makes motherboards, and Aerotek, which produces all sorts of radio equipment (Wi-Fi, mmWave, etc). These are small companies without the cachet of TSMC or MediaTek, but collectively they serve an important function in the market.Editor's Note:Guest author Jonathan Goldberg is the founder of D2D Advisory, a multi-functional consulting firm. Jonathan has developed growth strategies and alliances for companies in the mobile, networking, gaming, and software industries.The trouble here is that all of these companies were built on top of Intel. Look through their catalogs all those specialty servers and motherboards are built on Intel CPUs, often three years (or more) old.Prior to TSMC's rise as the flagship of Taiwan's electronics complex, Taiwanese companies formed the bedrock of modern compute, filling every available niche and spot in the supply chain. Over the years, many (but not all) of these companies moved their manufacturing base to the Mainland, but core R&D and headquarters remained in Taiwan. // Related StoriesNow all those companies face two major transitions. The first is they must weather Intel's turmoil. No matter what Intel's new CEO does with their fabs, the company is almost certain to face a major headcount reduction and a corresponding narrowing of their scope of business. This will very likely lead to a reduction in support for this ecosystem.Companies that today bridge the gap in providing support to the long tail of enterprise electronics will likely find it much harder to maintain previous levels of support.Their second major transition is, of course, a big part of the cause of Intel's troubles the shift to AI compute and the rise of Nvidia.Nvidia has become the dominant force in data center compute applications, and as AI percolates into the edge, that long tail will need to source specialty compute hardware from someone. But it is not clear the traditional vendors in Taiwan will be able to support that.At MWC, the one refuge from the sea of AI-washing marketing was the ITRI booth, where almost no one talked about AI (at least not in explicit marketing materials).Nvidia is now in the process of building up the electronics ecosystem that provided so much support to Intel. Nvidia's GTC, and the upcoming Computex, are very much about Nvidia's efforts to find, train, and support partners who can fulfill their ever-growing book of business. That being said, it is unlikely that Nvidia will build up its partner network in the same way that Intel did.The world has changed a lot, and companies which today build semi-custom server boxes for traditional compute may not be the ones to build semi-custom, long-tail server boxes for AI. In our tour, we saw almost no support for AMD, let alone Nvidia. This complex is built on x86 and general compute the AI complex is going to look different.Nvidia's GTC, and the upcoming Computex, are very much about Nvidia's efforts to find, train, and support partners who can fulfill their ever-growing book of business.In some senses, we could see some of these companies just altering their designs a motherboard for a CPU is built with largely the same materials as a GPU board. But so much of this business extends beyond the basics of metal bending.Companies which have been building Intel-based servers for decades do not have the in-house experience with Nvidia new firmware, new pin compatibility schemes, new power and cooling requirements, etc. This sort of soft expertise helped these companies differentiate themselves from the pure manufacturing-focused box builders in the Mainland.With much of that expertise now rendered less critical to customers, whatever will replace it is very much green field a jump ball where the traditional vendors have no clear advantage.This does not mean that disaster is imminent, but it seems likely that a big transformation is starting soon. Intel's strength in this corner of the industry, not only with the small specialty makers but their larger peers (and often parent companies), has long served as a major competitive advantage that tripped up other aspiring CPU vendors.Intel's ability to hold on to this complex will play a large role in determining its future. If Intel can remain the preferred silicon partner here, this segment will provide an important source of demand on which Intel can rebuild. But it is not clear how well these companies can navigate the changes coming in the next few years.

In context: The introduction of holographic displays marks a significant evolution in PC customization. While RGB lighting has been a staple for years, Coolify's Holo Fans elevate personalization to new heights by replacing static lighting effects with dynamic holograms. Coolify, a company specializing in PC customization, has developed a product that integrates holographic displays into desktop setups, potentially transforming the aesthetics of personal computing. The Holo Fans bring animated holograms to PC case fans, offering a visual experience that goes beyond traditional RGB lighting effects and provides a new level of personalization for gaming and workstation setups.Unlike conventional RGB fans that rely solely on LED lighting, Coolify's Holo Fans use spinning LED arrays to simulate holograms. At the center of these spinning fans, users will see floating holograms created by an array of 96 LEDs. These holograms are not merely static images; they can display logos, animations, and customizable graphics.Available in both 120mm and 140mm sizes, the Holo Fans come with a controller for managing animations and lighting effects. They also support software integration, allowing users to synchronize visuals with system metrics such as CPU temperature, fan speeds, and gaming performance indicators. The holograms can also be programmed to display system alerts or real-time notifications.The fans boast high-static pressure designs for efficient airflow, with speeds reaching up to 2,600 RPM. They deliver airflow rates of 52 CFM (Cubic Feet per Minute) and a rated air pressure of 2.75 mmH2O.Noise levels are kept at 33 dB(A), making them suitable for gaming and professional environments. However, their power consumption ranges between 2W and 6W and can peak at 10W higher than standard case fans but arguably justified by their features. // Related StoriesThe fans operate at fixed speeds, either 1,500 or 2,600 RPM, to maintain the holographic projection's integrity varying the speed would disrupt the visuals.One notable feature is how users interact with these fans. Unlike traditional PC peripherals that transfer data directly from the computer, Coolify's Holo Fans require images and animations to be uploaded via Wi-Fi using a companion smartphone app. While this may add some complexity during installation, especially for setups involving multiple fans, it eliminates concerns about resource-heavy PC software interfering with gaming performance.Originally unveiled at Computex 2024, the Holo Fans are now available for $45.99 each. However, due to high demand, early adopters may experience limited availability.

Something to look forward to: It's been eight years since Nintendo unleashed the Nintendo Switch on the world. While there have been several variants of that model since then, the company is finally unveiling the official successor in a few hours, and you can watch the whole thing live right here at 6am PDT / 9am ET. There have been a slew of leaks and rumors regarding the Switch 2 over the last 12 months or so. Nintendo offered some actual details during the first-look trailer in January, when it showed off the larger screen, top-mounted USB-C port, and an upgraded kickstand. We also saw the snazzy new controllers that appear to have PC-style mouse functionality.Following the first Nintendo Direct last week (below) that focused on new games, remasters, and the Virtual Game Cards feature that allows users to loan digital games to friends, the second Nintendo Direct takes place today.Nintendo is spending an hour on the Switch 2 reveal, which will hopefully mean we'll get an in-depth look at the console, its hardware, and all the features.There will also be several Switch 2 games unveiled. There was a brief glimpse of Mario Kart 9 during January's teaser, so expect to see plenty more details about that one later today.Other games that could be on display include a new 3D Mario, cross-gen titles, and some Microsoft and Xbox games. Microsoft Gaming CEO Phil Spencer said in January that he was looking forward to supporting Nintendo with the company's Xbox games. // Related StoriesIt's rumored that Halo: The Master Chief Collection, Forza Horizon 5, and Microsoft Flight Simulator 2024 are all heading to Nintendo's new handheld hybrid, and Microsoft signed a ten-year deal to bring Call of Duty to Nintendo's consoles after it acquired Activision Blizzard. Microsoft's Minecraft is already on the Switch, as are Pentiment and Grounded.There will be plenty of Switch 2 games shown off this week as Nintendo has announced the return of Nintendo Treehouse. These presentations take place at 7am PT / 10am ET today (straight after the Switch 2 livestream) and the same time tomorrow (April 4), when you can watch some hands-on gameplay of Switch 2 titles.

The big picture: 2025 is shaping up to be a pivotal year for Intel's comeback strategy. With a new CEO at the helm, the company is positioning its next-generation hardware as a litmus test for its return to the forefront of semiconductor manufacturing. As Intel's production roadmap reaches a critical juncture, how prospective clients will respond remains uncertain. Kevin O'Buckley, senior vice president and general manager of Intel Foundry Services, confirmed that risk production has begun for the company's upcoming 18A semiconductor node. The announcement, made at Intel's Vision conference, is likely to reassure investors and clients that development of the next generation of laptop and server processors has entered a crucial phase.Risk production, an industry-standard term, refers to the stage where manufacturers refine the production process for new chips. O'Buckley explained that Intel is currently scaling up from producing hundreds of 18A wafers at a time to thousands.Also see: How CPUs are Designed, Part 3:Building the ChipAlthough Intel has not yet named any external clients committed to building commercial chips on 18A, the company aims to reach volume and mass production in time to launch its Panther Lake CPUs later this year. With 28A tape-out design finalizations set to begin in the first half of 2025, Panther Lake is expected to deliver significant improvements in AI performance over Intel's Core Ultra 200V notebook processors.Intel's new CEO, Lip-Bu Tan, also recently confirmed that Nova Lake and Clearwater Forest chips are scheduled for release in 2026. Nova Lake will incorporate silicon from TSMC, Intel's foundry rival, while Clearwater Forest will bring 18A to the server market. // Related StoriesWith 18A, Intel is racing to outpace TSMC in bringing sub-3nm technologies to market specifically gate-all-around (GAA) transistors and backside power delivery. These innovations improve performance by reducing power leakage and enabling higher transistor density. TSMC does not plan to introduce GAA and backside power delivery until its N2 and A16 nodes debut next year. N2 is expected to enter early production soon and ramp up to mass production later this year.Intel's manufacturing has trailed behind TSMC and Samsung for several years. Following a string of disappointing quarters that led to the departure of former CEO Pat Gelsinger, the 18A node is seen as a crucial opportunity to reestablish confidence in Intel's foundry business.Nvidia and Broadcom are reportedly exploring the use of 18A for future products, though they are still in the early testing phase with Intel's wafers. Meanwhile, Apple is expected to be the first to adopt TSMC's N2 node, likely debuting it in the A20 processor for the iPhone 18 Pro in late 2026. AMD, Broadcom, Amazon AWS, and Intel are also expected to use N2.Nvidia and Broadcom are reportedly interested in using 18A for future products, but the two companies are likely just running early tests with Intel's wafers. Meanwhile, Apple is first in line for TSMC's N2, which will likely debut with the iPhone 18 Pro's A20 processor in late 2026. AMD, Broadcom, Amazon AWS, and Intel are also expected to employ N2.

The big picture: Chrome is by far the most popular browser across various platforms and device types, with a market share of over 66 percent. Needless to say, every change Google makes to its proprietary browser inevitably influences all major players in the web ecosystem. Google recently announced two major initiatives aimed at enhancing web security, with the ultimate goal of making encryption and certificate management more reliable and resilient against cybercrime. These new features are part of the Chrome Root Program, which, according to Google, demonstrates the company's commitment to strengthening online security through its Chrome browser.As the world's most popular browser vendor, Google is highly motivated to improve web safety by encouraging industry organizations to adopt its proposed standards. The latest changes from the search and advertising giant involve the CA/Browser Forum, a cross-industry group that establishes baseline requirements for issuing TLS certificates.TLS connections, which enable encrypted HTTPS protocols, are the backbone of modern web security. However, cybercriminals are constantly seeking ways to circumvent these protections. To counter this, Google has proposed two key measures: Multi-Perspective Issuance Corroboration (MPIC) and an automated vetting process known as "linting."Google explained that MPIC enhances existing methods for validating domain legitimacy before a Certificate Authority issues a new TLS certificate. The current process, known as "domain control validation," can be exploited in various ways, potentially leading to fraudulent certificate issuance. MPIC aims to mitigate these risks by introducing additional verification perspectives.The CA/Browser Forum unanimously adopted MPIC in a recent voting session, making it a mandatory requirement for Certificate Authorities during the certificate issuance process. Google also highlighted the Open MPIC Project as a robust implementation of this new validation method.While MPIC helps prevent fraudulent certificates from being issued, linting provides an additional layer of security by analyzing X.509 certificates for potential issues. The X.509 standard defines the format for public key certificates and plays a crucial role in the TLS protocol. With linting, CAs can verify whether a certificate is properly formatted for its intended use, such as website authentication. // Related StoriesLinting also identifies insecure certificates that rely on weak or outdated encryption technologies, thereby enhancing security and ensuring better interoperability between CAs through adherence to industry standards. Google noted that the linting process can be implemented through various open-source projects, including certlint, pkilint, x509lint, and zlint. The company once again secured unanimous support for linting in a recent CA/Browser Forum vote, and the technology officially became a requirement for new public certificates issued by CAs on March 15, 2025.

Why it matters: Vaping has been around for a long time, but its popularity didn't skyrocket until the early 2010s. Since then, it has significantly contributed to the world's e-waste problem, with thousands of disposable devices filling landfills and recycling centers. Not only are e-cigarettes damaging to the environment, but they are also a growing fire hazard. Fire Rover, a company that specializes in automated and semi-automated fire suppression systems, released its annual report noting that waste and recycling fires are steadily rising. In 2024, the company recorded 2,910 fires a 60-percent increase from 2023's 1,809 and a 100-percent jump from 2022's 1,409 incidents. The report also notes that fire crews dispatched to emergencies at trash and recycling facilities hit a record high of 398, a steady growth since Fire Rover began tracking the stat at 275 incidents in 2016.Lithium-ion battery fires are not new, nor are they the only cause of trash and recycling blazes. Fire Rover CEO Ryan Fogelman told Ars Technica things like fireworks, pool chemicals, and hot barbeque briquettes pose just as much risk. However, batteries, particularly those in disposable vaping products, are a rapidly growing cause mainly because of consumer ignorance and a lack of widespread e-waste collection.Many well-meaning customers know not to throw their vapes in the regular trash, so they use the other option recycling bins, which is no better. No matter which facility these devices land in, they can ignite in many ways. Crushing pressure, puncturing, short-circuiting, and vibration from facility operations are common causes. However, battery defects, internal cell failure, and overheating are indirect means of ignition that refuse centers cannot control. Fogelman estimates that about half of the incidents Fire Rover tracks are battery-related, costing facilities approximately $2.5 billion in 2024 alone.The CEO says that a properly functioning e-waste infrastructure could reduce this trend, but that does not currently exist and does not seem to be a high priority. Furthermore, the few facilities offering e-waste collection are abandoning or restricting it, likely because of the associated costs.For example, my local refuse center used to pick up e-waste once a year. It recently discontinued that service. Customers can still bring in their e-waste, but the facility has a long list of items it refuses to accept. The added inconvenience of having to haul in their old electronics and the annoyance of not having anywhere else to dispose of the unaccepted items has likely led many to just chuck the lot into the regular recycle bin or the trash. // Related StoriesVideo of a Fire Rover suppression system identifying fire and then smoke to extinguish a blaze before it spreads.Fire Rover points its finger at the vaping industry, believing it should take more responsibility for helping clean up the mess it has helped create."Not only are their batteries being improperly discarded in waste and recycling bins, but the vape industry has done the bare minimum to invest in the technology needed to address the 1.2 billion vapes entering our waste and recycling streams annually," the report states.Of course, you could always just use them to charge your phone.Image credit: Vaping360

What just happened? West Japan Railway Company (JR West) has unveiled what it claims to be the world's first 3D-printed train station building. Located at Hatsushima Station on the JR Kisei Main Line in Arida, Wakayama Prefecture, the project merges cutting-edge technology with cultural sensitivity, marking a potential milestone in railway infrastructure. The new station building, set to open in July, replaces an aging wooden structure that had served the rural community for years. Standing 2.6 meters tall, 6.3 meters wide, and 2.1 meters deep, the single-story reinforced concrete building was constructed using advanced 3D printing technology.Japanese 3D-printing construction specialist Serendix prefabricated four key components, including the roof and walls, off-site. These were then transported and assembled in just two and a half hours, during the window between the last train at night and the first train the following morning.The process involved shaping mortar with a 3D printer and creating formwork to mold the exterior. Reinforcing steel was inserted into the hollow sections before being filled with concrete, ensuring durability and earthquake resistance comparable to traditional reinforced concrete structures, without the need for conventional wooden or metal formwork.JR West estimates this approach cuts costs by around 50 percent compared to conventional reinforced concrete buildings.Beyond cost savings, this rapid construction method reduces disruptions to railway operations, making it an ideal solution for replacing aging infrastructure on active rail lines.The station's design reflects its local surroundings, featuring artistic depictions of Arida City's iconic mikan oranges and tachiuo (beltfish) on the exterior. JR West emphasized that the goal is to create a station cherished by residents while aligning with regional culture, an approach they describe as "regional coexistence." // Related StoriesHatsushima Station was chosen for the pilot due to its coastal location, allowing experts to evaluate the structure's resilience against environmental challenges such as salty air exposure. Additionally, JR West will assess long-term maintenance costs and feasibility as part of its broader plan to modernize Japan's railway infrastructure.If successful, JR West plans to expand 3D printing technology across its network. The company sees this innovation as key to its long-term vision of providing "safe, secure, and human-friendly transportation" while enhancing sustainability.Image credit: JIJI

Editor's take: Judging from the discourse in my communication bubble, almost no one likes AI technologies. The most enthusiastic AI evangelists are vendors interested in selling you something AI related to cash in on the craze or the companies creating it and shoving it down our throats. A recent Gartner analysis reports that spending on generative AI (GenAI) will grow to unprecedented levels in 2025, even though very few organizations understand how this technology can help their businesses or profit prospects. Experts believe this market "paradox" trend will persist through 2025 and 2026.Global spending on GenAI will reach around $644 billion in 2025, a 76.4-percent increase compared to 2024. This spending trend comes despite the technology's lack of actual usefulness and error rate. Gartner VP Analyst John-David Lovelock says foundational model providers are investing billions in developing and enhancing their solutions regardless of the growing concerns about the technology.This year, internal projects with lofty goals will face increased scrutiny, Lovelock stated. Chief investment officers will likely opt for commercial, off-the-shelf solutions, while proof-of-concept experimentations and self-developed implementations should become the exception. GenAI prospects haven't been as ambitious lately, but the technology should still significantly impact all IT spending this year.Experts expect AI technologies to become integral to all kinds of business operations and consumer products. This forced integration will be primarily due to AI capabilities baked directly into consumer hardware, with servers, smartphones, and PCs absorbing 80 percent of the projected GenAI spending.Lovelock notes that the growing prevalence of AI-enabled devices is driving market growth, with these consumer electronics expected to make up nearly the entire market by 2028. However, end consumers don't seem as enthusiastic about GenAI, as they're not actively seeking these "extra" features when shopping for new devices. Regardless, manufacturers continue embedding AI as a standard in mass-market products, ultimately forcing consumers to adopt it.Developers have adopted a service model that says, "You will consume our product, even if you don't want it." It sounds like a page out of Microsoft's playbook back in the day.

What just happened? Some significant changes are coming to Thunderbird. Mozilla, which financially resurrected the free, open-source email client a few years ago, will soon unveil a new set of "professional" services designed to enhance the basic email management features of the program. In 2023, Mozilla developers announced they were "remaking" Thunderbird's software architecture, aiming to modernize its crumbling and unsustainable codebase. The company later confirmed plans to add new paid professional accounts available for all Thunderbird users.Mozilla Managing Director Ryan Sipes said they have named the new bundle of services "Thunderbird Pro," which includes several new features. Meanwhile, Mozilla is rebranding the email client to "Thundermail." The Pro package expands and enhances what Thunderbird has to offer its users.At first, the company will offer Thunderbird Pro for free "to consistent community contributors." Everyone else will have to pay for the pro tier. However, once Thunderbird Pro has built a sustainable user base, Mozilla will consider adding limited free tiers.Thunderbird was losing users regularly to "rich" ecosystems with local clients and remote services. Gmail and Office 365 are the main competitors. While these platforms lock their customers into their ecosystems, Thunderbird Pro will supposedly offer a 100-percent open source, freedom-respective alternative for those seeking and all-in-one experience.Thunderbird Pro will include the following services:Thunderbird AppointmentThunderbird SendThunderbird AssistThundermailThunderbird Appointment is a scheduling tool that allows users to send a link to someone as an invitation to a calendar appointment. The service should make meeting with other people more streamlined, Sipes noted, adding that he and his team weren't happy with the currently available tools. // Related StoriesThunderbird Send is a spiritual rebirth of Firefox Send, the file-sharing service Mozilla shut down in 2020. The development team rebuilt the project to offer more direct methods of sending files.Sipes described Thunderbird Assist as an experimental AI service. He was vague about the details, so it may still be in early development. He said users would have access to some unnamed AI features remotely processed by Flower AI. More powerful devices will have the option to run AI models locally.Finally, Thundermail is the "new" email client akin to Gmail or Outlook. Sipes believes the revamped service can provide a better service than any of its competitors while adhering to Mozilla's core values. Thundermail is built on the open-source software stack of Stalwart, improving its basic capabilities as well.