The Wall Street Journal profiles "the startup that is now one of a handful of chip makers nipping at the heels of Nvidia." Furiosa's AI chip is dubbed "RNGD" — short for renegade — and slated to start mass production this month. Valued at nearly $700 million based on its most recent fundraising, Furiosa has attracted interest from big tech firms. Last year, Meta Platforms attempted to acquire it, though the startup declined the offer. OpenAI used a Furiosa chip for a recent demonstration in Seoul. LG's AI research unit is testing the chip and said it offered "excellent real-world performance." Furiosa said it is engaged in talks with potential customers.

Nvidia's graphic processing units, or GPUs, dominated the initial push to train AI models. But companies like Furiosa are betting that for the next stage — referred to as "inference," or using AI models after they're trained — their specialty chips can be competitive. Furiosa makes chips called neural processing units, or NPUs, which are a rising class of chips designed specifically to handle the type of computing calculations underpinning AI and use less energy than GPUs. [Founder/CEO June] Paik said Furiosa's chips can provide similar performance as Nvidia's advanced GPUs with less electricity usage. That would drive down the total costs of deploying AI. The tech world, Paik says, shouldn't be so reliant on one chip maker for AI computing. "A market dominated by a single player — that's not a healthy ecosystem, is it?" Paik said...

In 2024, at Stanford's prestigious Hot Chips conference, Paik debuted Furiosa's RNGD chip as a solution for what he called "sustainable AI computing" in a keynote speech. Paik presented data showing how the chip could run the then-latest version of Meta's Llama large language model with more than twice the power efficiency of Nvidia's high-end chips. Furiosa's booth was swarmed with engineers from big tech firms, including Google, Meta and Amazon.com, wanting to see a live demo of the chip. "It was a moment where we felt we could really move forward with our chip with confidence," Paik said.

The New York Times checks in on U.S. university researchers and start-ups trying to create domestic rare-earth processing facility: There is too little money to be made in rare earths for the elements to be of much interest to mining giants, so the challenge of reestablishing a domestic industry has fallen to small companies like Phoenix Tailings, a Boston-area startup that runs the metal-making plant in Exeter, New Hampshire. A handful of other companies in the United States are processing rare earths in small quantities, including MP Materials, which owns a mine in Mountain Pass, California, and recently began producing rare-earth metal in Fort Worth, Texas. Similar efforts are underway in Europe and Asia. "It's small volumes of low-value materials that are very expensive to process," said Elsa Olivetti, a materials science and engineering professor at the Massachusetts Institute of Technology. "Meaning it's hard to make money."

Phoenix Tailings' New Hampshire operation is about 2 months old, housed in a converted medical device plant. The company buys metric-ton bags of powder — a mixture of neodymium and praseodymium bound with oxygen — from mining and refining companies in the United States, South America and Australia. It funnels that flour-like material into a drying oven and eventually into furnaces that heat it to the temperature of volcanic lava. This circuit takes up less than 15,000 square feet and is designed to generate no emissions other than those associated with the electricity Phoenix Tailings uses. The closed-loop design distinguishes this process from the more energy-intensive techniques used in China, where workers scoop up molten metal with ladles. That approach releases perfluorocarbons, potent greenhouse gases that do not break down easily.
In late 2024 the company was three weeks from bankruptcy — but it's recently been valued at $189 million.

An anonymous reader shared this report from Electrek: Tesla has released a new video showing a Tesla Semi truck charging at a massive 1.2 megawatts (MW), finally giving us a clear look at the charging speeds that will enable long-haul electric trucking...
>
Tesla claimed the Semi would be able to charge 70% of its range in 30 minutes. For a truck with a 500-mile range and an estimated battery pack of around 800-900 kWh, that requires an incredibly high power output, well beyond the 250 kW or even 350 kW we see on passenger EVs in North America. Today, the official Tesla Semi account on X released a video showing exactly that. In the video, Tesla engineers are seen monitoring a charging session where the power output climbs to a peak of 1.2 MW (1,206 kW).

This is consistent with the capabilities Tesla announced for its new V4 Cabinet architecture earlier this year. The V4 cabinets are designed to support 400V-1000V vehicle architectures and can deliver up to 500 kW for cars (like the Cybertruck) and up to 1.2 MW for the Semi. There is some information missing from the video. For example, we don't see the state-of-charge of the truck, so we don't at what battery percentage Tesla Semi can achieve and maintain this charge rate. Peak speed is one thing, but sustaining that power without overheating the pack or the cable is the real challenge. The liquid-cooled charging cable and the immersion-cooled connector (part of the Megawatt Charging System or a high-power proprietary Tesla solution, though Tesla has been leaning toward MCS compatibility) seem to be doing their job....

This comes just as Tesla is gearing up for volume production of the Semi at its new factory expansion near Gigafactory Nevada. The automaker is targeting a start of production in the first half of 2026 and a ramp up to volume production in the second half.

Pibroch(CiH) writes: According to the obituary linked, Stewart Cheifet of Computer Chronicles fame has died. The obituary states he passed Dec 28, 2025. Cheifet and Digital Research founder Gary Kildall hosted the public television show The Computer Chronicles starting in 1984, and Stewart continued to host the show well into the 1990s. He was well-known for his affable presence and adeptness at interviewing guests and finding out the straight dope about their products. He had recently undergone spinal surgery and had somewhat disappeared from public view after the death of his wife Peta in 2024.

Longtime Slashdot reader ptorrone writes: The January 1, 2026, NYC mayoral inauguration prohibits attendees from bringing specific brand-name devices, explicitly banning Raspberry Pi single-board computers and the Flipper Zero, listed alongside weapons, explosives, and drones. Rather than restricting behaviors or capabilities like signal interference or unauthorized transmitters, the policy names two widely used educational and testing tools while allowing smartphones and laptops that are far more capable. Critics argue this device-specific ban creates confusion, encourages selective enforcement, and reflects security theater rather than a clear, capability-based public safety framework. New York has handled large-scale events more pragmatically before.

An anonymous reader quotes a report from the New York Times: South Africans ... have found a remedy for power cuts that have plagued people in the developing world for years. Thanks to swiftly falling prices of Chinese made solar panels and batteries, they now draw their power from the sun. These aren't the tiny, old-school solar lanterns that once powered a lightbulb or TV in rural communities. Today, solar and battery systems are deployed across a variety of businesses -- auto factories and wineries, gold mines and shopping malls. And they are changing everyday life, trade and industry in Africa's biggest economy. This has happened at startling speed. Solar has risen from almost nothing in 2019 to roughly 10 percent of South Africa's electricity-generating capacity.

No longer do South Africans depend entirely on giant coal-burning plants that have defined how people worldwide got their electricity for more than a century. That's forcing the nation's already beleaguered electric utility to rethink its business as revenues evaporate. Joel Nana, a project manager with Sustainable Energy Africa, a Cape Town-based organization, called it "a bottom-up movement" to sidestep a generations-old problem. "The broken system is unreliable electricity, expensive electricity or no electricity at all," he said. "We've been living in this situation forever." What's happening in South Africa is repeating across the continent. Key to this shift: China's ambition to lead the world in clean energy. The report says that more than 7 gigawatts of solar capacity have been installed in South Africa over the past five years -- about 1/10 of the country's total installed capacity (55 GW). And most of this new solar capacity is privately owned and installed by households and businesses rather than utilities.

Across the continent, Chinese solar imports rose 50% in the first 10 months of 2025. Cheap Chinese solar is rapidly reshaping Africa's energy landscape from the bottom up but it's also shifting geopolitical influence, hollowing out local manufacturing opportunities, and deepening divides between those who can afford energy independence and those who can't. "The solar surge does little to address the most pressing social and economic problems of developing countries like South Africa, the need to generate new jobs for millions of young citizens," reports the NYT. "Installation labor is local, but the panels and batteries are almost all made in China."

Further reading: Why Solarpunk Is Already Happening In Africa

One-Netbook has unveiled the OneXSugar Wallet, the first gaming handheld with a folding OLED display. The Verge reports: The OneXSugar Wallet was announced on China's Weibo yesterday, but with few details about its features and capabilities. That folding OLED screen has a resolution of 2480 x 1860 pixels, and the handheld will be powered by an unspecified "Qualcomm gaming platform flagship processor," but its performance and emulation capabilities are unknown.

Based on photos and a video released by One-Netbook, the OneXSugar Wallet will feature a standard set of controls including asymmetrical thumbsticks, four action buttons, and a D-pad situated on either side of the lower half of its display. There are also shoulder buttons and triggers on the back of the handheld, and a pair of front-facing speakers flanking the top half of the screen. The biggest question is how much will the handheld cost...

An anonymous reader quotes a report from Ars Technica: The nervous system does an astonishing job of tracking sensory information, and does so using signals that would drive many computer scientists insane: a noisy stream of activity spikes that may be transmitted to hundreds of additional neurons, where they are integrated with similar spike trains coming from still other neurons. Now, researchers have used spiking circuitry to build an artificial robotic skin, adopting some of the principles of how signals from our sensory neurons are transmitted and integrated. While the system relies on a few decidedly not-neural features, it has the advantage that we have chips that can run neural networks using spiking signals, which would allow this system to integrate smoothly with some energy-efficient hardware to run AI-based control software.

[...] There are four ways that these trains of spikes can convey information: the shape of an individual pulse, through their magnitude, through the length of the spike, and through the frequency of the spikes. Spike frequency is the most commonly used means of conveying information in biological systems, and the researchers use that to convey the pressure experienced by a sensor. The remaining forms of information are used to create something akin to a bar code that helps identify which sensor the reading came from. In addition to registering the pressure, the researchers had each sensor send a "I'm still here" signal at regular time intervals. Failure to receive this would be an indication that something has gone wrong with a sensor.

The spiking signals allow the next layer of the system to identify any pressure being experienced by the skin, as well as where it originated. This layer can also do basic evaluation of the sensory input: "Pressure-initiated raw pulses from the pulse generator accumulated in the signal cache center until a predefined pain threshold is surpassed, activating a pain signal." This can allow the equivalent of basic reflex reactions that don't involve higher-level control systems. For example, the researchers set up a robotic arm covered with their artificial skin, and got it to move the arm whenever it experiences pressure that can cause damage. The second layer also combines and filters signals from the skin before sending the information on to the arm's controller, which is the equivalent of the brain in this situation. So, the same system caused a robotic face to change expressions based on how much pressure its arm was sensing.

[...] The skin is designed to be assembled from a collection of segments that can snap together using magnetic interlocks. These automatically link up any necessary wiring, and each segment of skin broadcasts a unique identity code. So, if the system identifies damage, it's relatively easy for an operator to pop out the damaged segment and replace it with fresh hardware, and then update any data that links the new segment's ID with its location. The researchers call their development a neuromorphic robotic e-skin, or NRE-skin. "Neuromorphic" as a term is a bit vague, with some people using it to mean a technology that directly follows the principles used by the nervous system. That's definitely not this skin. Instead, it uses "neuromorphic" far more loosely, with the operation of the nervous system acting as an inspiration for the system.The findings have been published in the journal PNAS.

Amid a global DDR5 shortage and soaring prices, Russian hardware enthusiasts are experimenting with do-it-yourself DDR5 RAM by sourcing empty PCBs and soldering memory chips by hand. Tom's Hardware reports: The idea comes from Russian YouTuber PRO Hi-Tech's Telegram channel, where a local enthusiast known as "Vik-on" already performs VRAM upgrades for GPUs, so this is a relatively safe operation for him. According to Vik-on, empty RAM PCBs can be sourced from China for as little as $6.40 per DIMM. The memory chips themselves, though, that's a different challenge.

The so-called spot market for memory doesn't really exist at the moment, since no manufacturer has the production capacity to make more RAM, and even if they did, they'd sell to better-paying AI clients instead. Still, you can find SK Hynix and Samsung chips across Chinese marketplaces if you search for the correct part number, as shown in the attached screenshots.

Moreover, the Telegram thread says it would cost roughly 12,000 Russian Rubles ($152) to build a 16 GB stick with "average" specs, which is about the same as a retail 16 GB kit. There's also a ZenTimings snapshot showing CL28 timings, claiming that even relatively high-end DDR5 RAM can be built using this method, but it won't be cost-effective. Therefore, it doesn't make too much sense just yet to get the BGA rework station out and assemble your own DDR5. Things are expected to get worse, though, so maybe these Russians are on to something.

Even after its acquisition by Qualcomm, the EFF believes Arduino "isn't imposing any new bans on tinkering with or reverse engineering Arduino boards," (according to Mitch Stoltz, EFF director for competition and IP litigation). While Adafruit's managing editor Phillip Torrone had claimed to 36,000+ followers on LinkedIn that Arduino users were now "explicitly forbidden from reverse engineering," Arduino corrected him in a blog post, noting that clause in their Terms & Conditions was only for Arduino's Software-as-a-Service cloud applications. "Anything that was open, stays open."

And this week EE Times spoke to Guneet Bedi, SVP of Arduino, "who was unequivocal in saying that Arduino's governance structure had remained intact even after the acquisition." "As a business unit within Qualcomm, Arduino continues to make independent decisions on its product portfolio, with no direction imposed on where it should or should not go," Bedi said. "Everything that Arduino builds will remain open and openly available to developers, with design engineers, students and makers continuing to be the primary focus.... Developers who had mastered basic embedded workflows were now asking how to run large language models at the edge and work with artificial intelligence for vision and voice, with an open source mindset," he said. According to Bedi, this was where Qualcomm's technology became relevant. "Qualcomm's chipsets are high performance while also being very low power, which comes from their mobile and Android phone heritage. Despite being great technology, it is not easily accessible to design engineers because of cost and complexity. That made this a strong fit," he said.

The most visible outcome of this acquisition is Uno Q, which Bedi described as being comparable to a mid-tier Android phone in capability, starting at a price of $44. For Arduino, this marked a shift beyond microcontrollers without abandoning them. "At the end of the day, we have not gone away from our legacy," Bedi said. "You still have a real-time microcontroller, and you still write code the way Arduino developers are used to. What we added is compute, without forcing people to change how they work." Uno Q combines a Linux-based compute system with a real-time microcontroller from the STM32 family. "You do not need two different development environments or two different hardware platforms," Bedi added... Rather than introducing a customized operating system, Arduino chose standard Debian upstream. "We are not locking developers into anything," Bedi said. "It is standard Debian, completely open...." Pre-built models covering tasks like object detection and voice recognition run locally on the board....

While the first reference design uses Qualcomm silicon, Bedi was careful to stress that this does not define the roadmap. "There is zero dependency on Qualcomm silicon," he said. "The architecture is portable. Tomorrow, we can run this on something else." That distinction matters, particularly for developers wary of vendor lock-in following the acquisition. Uno Q does compete directly with platforms like Raspberry Pi and Nvidia Jetson, but Bedi framed the difference less in terms of raw performance and more in flexibility. "When you build on those platforms, you are locked to the board," he said. "Here, you can build a prototype, and if you like it, you can also get access to the chip and design your own hardware." With built-in storage removing the need for external components, Uno Q positions itself less as a faster board and more as a way to simplify what had become an increasingly messy development stack...

Looking a year ahead, Bedi believes developers should experience continuity rather than disruption. The familiar Arduino approach to embedded and real-time systems remains unchanged, while extending naturally into more compute-intensive applications... Taken together, Bedi's comments suggest that Arduino's post-acquisition direction is less about changing what Arduino is, and more about expanding what it can realistically be used for, without abandoning the simplicity that made it relevant in the first place.
"We want to redefine prototyping in the age of physical artificial intelligence," Bedi said...

Laptop makers are facing record-setting memory prices next year. The site Notebookcheck catalogs how different companies are responding: Sources told [Korean business newspaper] Chosun Biz that some manufacturers have signed preliminary contracts with Samsung, Micron, and SK Hynix. Even so, it won't prevent DDR5 RAM prices from soaring 45% higher by the end of 2026.... Before the memory shortage, PC sales had been on the upswing in part because of forced Windows 11 upgrades. That trend will likely reverse in 2026, as buyers avoid Lenovo laptops and alternatives from its rivals.

Realizing a slowdown in purchases is inevitable, postponed launches are one potential outcome. Other manufacturers, including Dell and Framework have already announced impending price hikes... [The article also cites reports that one laptop manufacturer "plans to raise the prices of high-end models by as much as 30%."] U.S.-based Maingear now encourages customers to mail in their own modules to complete custom builds. Yet, without recycling parts from older systems, that won't result in significant savings for consumers.

An anonymous Slashdot reader shared this story from Interesting Engineering: Cybersecurity specialists from the research group DARKNAVY have demonstrated how modern humanoid robots can be compromised and weaponised through weaknesses in their AI-driven control systems.

In a controlled test, the team demonstrated that a commercially available humanoid robot could be hijacked with nothing more than spoken commands, exposing how voice-based interaction can serve as an attack vector rather than a safeguard, reports Yicaiglobal... Using short-range wireless communication, the hijacked machine transmitted the exploit to another robot that was not connected to the network. Within minutes, this second robot was also taken over, demonstrating how a single breach could cascade through a group of machines. To underline the real-world implications, the researchers issued a hostile command during the demonstration. The robot advanced toward a mannequin on stage and struck it, illustrating the potential for physical harm.

The 2011 meltdown at Fukushima's nuclear plant "was the world's worst nuclear disaster since Chernobyl in 1986," CNN remembers.

But this week Japanese authorities "have approved a decision to restart the world's biggest nuclear power plant," reports CNN, "which has sat dormant for more than a decade following the Fukushima nuclear disaster."

Despite nerves from many local residents, the Niigata prefectural assembly, home to the Kashiwazaki-Kariwa plant, approved a bill on Monday that clears the way for utility company Tokyo Electric Power Company (TEPCO) to restart one of the plant's seven reactors. The company plans to bring the No. 6 reactor back online around January 20, Japan's public broadcaster NHK reported...

Following the [2011] disaster, Japan shut down all 54 of its nuclear power stations including Kashiwazaki-Kariwa, which sits in the coastal and port region of Niigata about 320 kilometers (200 miles) north of Tokyo on Japan's main island of Honshu. Japan has since restarted 14 of the 33 nuclear reactors that remain operable, according to the World Nuclear Association. The Niigata plant will be the first to reopen under the operation of TEPCO, the company that ran the Fukushima Daiichi power station. It has been trying to reassure residents of the restart plan is safe...

About 60-70% of Japan's power generation comes from imported fossil fuels, which cost the country about 10.7 trillion yen ($68 billion) last year alone... Japan is the world's fifth-largest emitter of carbon dioxide, after China, the United States, India and Russia, according to the International Energy Agency. But it has committed to reaching net zero emissions by 2050, and renewable energy was at the center of its latest energy plan published earlier this year, with a push for greater investments in solar and wind. The country's energy demands are also expected to increase in the coming years due to a boom in energy-hungry data centers that power AI infrastructure. To achieve its energy and climate goals, Japan aims to double the share of nuclear power in its electricity mix to 20% by 2040...

On its website, TEPCO said Kashiwazaki-Kariwa had undergone multiple inspections and upgrades and that the company had learned "the lessons of Fukushima." The company said new seawalls and watertight doors would provide "stronger protection against tsunamis" and that mobile generators and more fire trucks would be on hand for "cooling support" in an emergency. It also said the plant now had "upgraded filtering systems designed to control the spread of radioactive materials."
A survey published by the prefecture in October "found 60% of residents did not think conditions for the restart had been met," reports Reuters, adding that "Nearly 70% were worried about TEPCO operating the plant."

An anonymous reader quotes a report from the Financial Post: A Texas power developer is proposing to repurpose nuclear reactors from Navy warships to power the United States grid as the Trump administration pushes to secure massive amounts of energy for the artificial intelligence boom. HGP Intelligent Energy LLC filed an application to the Energy Department to redirect two retired reactors to a data center project proposed at Oak Ridge, Tennessee, according to a letter submitted to the agency's Office of Energy Dominance Financing. The project, filed for the White House's Genesis Mission, would produce about 450-520 megawatts of around-the-clock electricity, or enough to power roughly 360,000 homes. The proposal would rewire reactors from naval vessels, originally built by Westinghouse Electric Company and General Electric, at a fraction of the cost of new builds.

According to the report, The developer expects to seek a loan guarantee from the U.S. Department of Energy and raise roughly $1.8-$2.1 billion in private capital to prepare the reactors for civilian use, targeting initial completion by 2029. The approach is technically feasible but would break new ground by adapting military nuclear assets for the commercial grid. Bloomberg first reported the story.

An anonymous reader shares a report: In 2024, 25.2% of gross final energy consumption in the EU came from renewable sources, up by 0.7 percentage points compared with 2023. This share is 17.3 pp short of meeting the 2030 target (42.5%), which would require an annual average increase of 2.9 pp from 2025 to 2030.

Among the EU countries, Sweden recorded the highest share of its gross final energy consumption coming from renewable sources (62.8%). Sweden primarily relied on solid biomass, hydro and wind. Finland followed with 52.1%, relying on solid biomass, wind and hydro, while Denmark came in third with 46.8%, with most of its renewable energy sourced from solid biomass, wind and biogas. The lowest shares of renewables were recorded in Belgium (14.3%), Luxembourg (14.7%), and Ireland (16.1%).

In July Google promised to scale the CO2 batteries of "Energy Dome" as a long-duration energy storage solution. Now IEEE Spectrum visits its first plant in Sardinia, where 2,000 tonnes of carbon dioxide power a turbine generating 20 MW over 10 hours — storing "large amounts of excess renewable energy until it's needed..."

"Google likes the concept so much that it plans to rapidly deploy the facilities in all of its key data-center locations in Europe, the United States, and the Asia-Pacific region." Developed by the Milan-based company Energy Dome, the bubble and its surrounding machinery demonstrate a first-of-its-kind "CO2 Battery," as the company calls it... And in 2026, replicas of this plant will start popping up across the globe. We mean that literally. It takes just half a day to inflate the bubble. The rest of the facility takes less than two years to build and can be done just about anywhere there's 5 hectares of flat land.

The first to build one outside of Sardinia will be one of India's largest power companies, NTPC Limited. The company expects to complete its CO2 Battery sometime in 2026 at the Kudgi power plant in Karnataka, in India. In Wisconsin, meanwhile, the public utility Alliant Energy received the all clear from authorities to begin construction of one in 2026 to supply power to 18,000 homes... The idea is to provide electricity-guzzling data centers with round-the-clock clean energy, even when the sun isn't shining or the wind isn't blowing. The partnership with Energy Dome, announced in July, marked Google's first investment in long-duration energy storage...

CO2 Batteries check a lot of boxes that other approaches don't. They don't need special topography like pumped-hydro reservoirs do. They don't need critical minerals like electrochemical and other batteries do. They use components for which supply chains already exist. Their expected lifetime stretches nearly three times as long as lithium-ion batteries. And adding size and storage capacity to them significantly decreases cost per kilowatt-hour. Energy Dome expects its LDES solution to be 30 percent cheaper than lithium-ion.

China has taken note. China Huadian Corp. and Dongfang Electric Corp. are reportedly building a CO2-based energy-storage facility in the Xinjiang region of northwest China.
Google's senior lead for energy storage says they like how Energy Dome's solution can work in any region. "They can really plug and play this."

And they expect Google to help the technology "reach a massive commercial stage."

In 2023 Slashdot covered a battery-swapping startup that promised to give EVs a full charge in about the same time it takes to fill a tank of gas.

They just filed for bankruptcy, reports Inc: Ample was founded in 2014 with a goal of "solving slow charging times and infrastructure incompatibility" for commercial EV fleets such as those in logistics, ride-hailing, and delivery, the filing states. To-date, Ample has raised more than $330 million across five rounds of funding to finance research and development and deployment. Rather than tackling fast charging, its strategy involved developing "fully autonomous modular battery swapping," capable of delivering a fully charged battery in just five minutes. The technology requires purpose-built "Ample stations" that look a little like carwashes. A car is guided into the bay and elevated on a platform. A robot then identifies the location of a car's battery module, removes it, and replaces it with a charged module, Canary Media reported.

The company also boasts partnerships with Uber, Mitsubishi, and Stellantis, and notes it has deployed its technology — or is pursuing deployment — in San Francisco, Madrid and Tokyo. Even so, it ran up against funding issues. In its filing, Ample attributed its bankruptcy to macroeconomic and industry headwinds, such as "severe supply chain disruptions," "contraction in both public and private investment in renewable energy" and the "reduction, delay, or redirection of government incentives intended to accelerate EV adoption." The filing notes that regulatory and permitting delays slowed its launch in international markets, after which access to capital foiled its scaling efforts. The company eliminated all but two full-time, non-executive employees after formerly employing about 200...

Electrek noted that Ample is the second battery swapping startup to go bankrupt after California-based Better Place in collapsed in 2013 amid financial issues related to how capital intensive it was to build infrastructure, Reuters reported. And Tesla briefly pursued the concept, building a station in California, before ditching the idea altogether.
Ample "claimed to have designed autonomous battery swapping stations that would be rapidly deployable, cheap to build, and could adapt to any EV design with a modular battery which would be easy for manufacturers to use," notes Electrek's article: Where this bankruptcy leaves Ample's technology is unclear. Another company could snap it up and try to do something with it, if they find that the technology is real and useful. Ample had gotten investments and partnerships with Shell, Mitsubishi and Stellantis, for example, so the company wasn't alone in touting its tech. Or, it could just disappear, as other EV battery swapping plans have before...

That's not to say that nobody has been successful at at implementing battery swap, though. NIO seems to be successful with its battery swapping tech in China, though the company did miss its 2025 scaling goals by a longshot. But as of yet, this is the only notable example of a successful battery swap initiative, and it was done by an automaker itself, rather than a startup claiming to work for every automaker.
Electrek's writer is "just not bullish on battery swapping as a solution in general. Currently, the fastest-charging vehicles can charge from 10-80% in about 18 minutes. While that's longer than 5 minutes, it's not really a terrible amount of time to spend during most stops."

Plus, if cars come and go in 5 minutes instead of 18 minutes, "then you're going to have more than triple the throughput at peak utilization." And Ample's prices would be about the same as normal EV quick-charging prices...

An anonymous reader quotes a report from Ars Technica: At this point, most competitive online multiplayer games on the PC come with some kind of kernel-level anti-cheat software. As we've written before, this is software that runs with more elevated privileges than most other apps and games you run on your PC, allowing it to load in earlier and detect advanced methods of cheating. More recently, anti-cheat software has started to require more Windows security features like Secure Boot, a TPM 2.0 module, and virtualization-based memory integrity protection. Riot Games, best known for titles like Valorant and League of Legends and the Vanguard anti-cheat software, has often been one of the earliest to implement new anti-cheat requirements. There's already a long list of checks that systems need to clear before they'll be allowed to play Riot's games online, and now the studio is announcing a new one: a BIOS update requirement that will be imposed on "certain players" following Riot's discovery of a UEFI bug that could allow especially dedicated and motivated cheaters to circumvent certain memory protections.

In short, the bug affects the input-output memory management unit (IOMMU) "on some UEFI-based motherboards from multiple vendors." One feature of the IOMMU is to protect system memory from direct access during boot by external hardware devices, which otherwise might manipulate the contents of your PC's memory in ways that could enable cheating. The patch for these security vulnerabilities (CVE-2025-11901, CVE-202514302, CVE-2025-14303, and CVE-2025-14304) fixes a problem where this pre-boot direct memory access (DMA) protection could be disabled even if it was marked as enabled in the BIOS, creating a small window during the boot process where DMA devices could gain access to RAM.

The relative obscurity and complexity of this hardware exploit means that Vanguard isn't going to be enforcing these BIOS requirements on every single player of its games. For now, it will just apply to "restricted" players of Valorant whose systems, for one reason or another, are "too similar to cheaters who get around security features in order to become undetectable to Vanguard." But Riot says it's considering rolling the BIOS requirement out to all players in Valorant's highest competitive ranking tiers (Ascendant, Immortal, and Radiant), where there's more to be gained from working around the anti-cheat software. And Riot anti-cheat analyst Mohamed Al-Sharifi says the same restrictions could be turned on for League of Legends, though they aren't currently. If users are blocked from playing by Vanguard, they'll need to download and install the latest BIOS update for their motherboard before they'll be allowed to launch the game. Riot's new anti-cheat change could create problems for older PCs if the new anti-cheat change is expanded, notes Ars.

The update relies on a BIOS patch to fix a UEFI flaw, and many older motherboards, especially Intel 300-series and AMD AM4 boards, may never receive that update. If Riot flags a system and the manufacturer doesn't provide a patched BIOS, players could be locked out of games despite having otherwise capable hardware.

An anonymous reader quotes a report from Tom's Hardware: A North Korean imposter was uncovered, working as a sysadmin at Amazon U.S., after their keystroke input lag raised suspicions with security specialists at the online retail giant. Normally, a U.S.-based remote worker's computer would send keystroke data within tens of milliseconds. This suspicious individual's keyboard lag was "more than 110 milliseconds," reports Bloomberg. Amazon is commendably proactive in its pursuit of impostors, according to the source report.

The news site talked with Amazon's Chief Security Officer, Stephen Schmidt, about this fascinating new case of North Koreans trying to infiltrate U.S. organizations to raise hard currency for the Democratic People's Republic of Korea (DPRK), and sometimes indulge in espionage and/or sabotage. Schmidt says that Amazon has foiled more than 1,800 DPRK infiltration attempts since April 2024. Moreover, the rate of attempts continues apace, with Amazon reckoning it is seeing a 27% QoQ uplift in North Koreans trying to get into the Amazon corporation. However, Amazon's success can be almost entirely credited to the fact that it is actively looking for DPRK impostors, warns its Chief Security Officer. "If we hadn't been looking for the DPRK workers," Schmidt said, "we would not have found them."

A new study estimates that AI systems in 2025 consumed as much electricity as New York City emits in carbon pollution and used hundreds of billions of liters of water, driven largely by power-hungry data centers and cooling needs. Researchers say the real impact is likely higher due to poor transparency from tech companies about AI-specific energy and water use. "There's no way to put an extremely accurate number on this, but it's going to be really big regardless... In the end, everyone is paying the price for this," says Alex de Vries-Gao, a PhD candidate at the VU Amsterdam Institute for Environmental Studies who published his paper today in the journal Patterns. The Verge reports: To crunch these numbers, de Vries-Gao built on earlier research that found that power demand for AI globally could reach 23GW this year -- surpassing the amount of electricity used for Bitcoin mining in 2024. While many tech companies divulge total numbers for their carbon emissions and direct water use in annual sustainability reports, they don't typically break those numbers down to show how many resources AI consumes. De Vries-Gao found a work-around by using analyst estimates, companies' earnings calls, and other publicly available information to gauge hardware production for AI and how much energy that hardware likely uses.

Once he figured out how much electricity these AI systems would likely consume, he could use that to forecast the amount of planet-heating pollution that would likely create. That came out to between 32.6 and 79.7 million tons annually. For comparison, New York City emits around 50 million tons of carbon dioxide annually. Data centers can also be big water guzzlers, an issue that's similarly tied to their electricity use. Water is used in cooling systems for data centers to keep servers from overheating. Power plants also demand significant amounts of water needed to cool equipment and turn turbines using steam, which makes up a majority of a data center's water footprint. The push to build new data centers for generative AI has also fueled plans to build more power plants, which in turn use more water and (and create more greenhouse gas pollution if they burn fossil fuels).

AI could use between 312.5 and 764.6 billion liters of water this year, according to de Vries-Gao. That reaches even higher than a previous study conducted in 2023 that estimates that water use could be as much as 600 billion liters in 2027. "I think that's the biggest surprise," says Shaolei Ren, one of the authors of that 2023 study and an associate professor of electrical and computer engineering at the University of California, Riverside. "[de Vries-Gao's] paper is really timely... especially as we are seeing increasingly polarized views about AI and water," Ren adds. Even with the higher projection for water use, Ren says de Vries-Gao's analysis is "really conservative" because it only captures the environmental effects of operating AI equipment -- excluding the additional effects that accumulate along the supply chain and at the end of a device's life.