Tobias Mann reports via The Register: The land surrounding a nuclear power plant might not sound like prime real estate, but as more bit barns seek to trim costs, it's poised to become a rather hot commodity. All datacenters are energy-hungry but with more watt-greedy AI workloads on the horizon, nuclear power has fresh appeal, especially for hyperscalers. Such a shift in power also does wonders for greenwashing narratives around net-zero operations. While not technically renewable, nuclear power does have the benefit of being carbon-free, not to mention historically reliable -- with a few notable exceptions of course. All of these are purported benefits cited by startup NE Edge, which has been fighting for more than a year to be able to build a pair of AI datacenters adjacent to a 2GW Millstone nuclear power plant in Waterford, Connecticut.

According to the Hartford Courant, NE Energy has secured $1.6 billion to construct the switching station and bit barns, which will span 1.2 million square feet in total. NE Energy will reportedly spend an equivalent sum on between 25,000 and 35,000 servers. Considering the price of GPU systems from Nvidia, AMD, and Intel, we suspect that those figures probably refer to the number of GPUs. We've asked NE Edge for more information. NE Energy has faced local challenges getting the project approved because residents are concerned the project would end up increasing the cost of electricity. The facilities will reportedly consume as much as 13 percent of the plant's output. The project's president Thomas Quinn attempted to quell concerns, arguing that by connecting directly to the plants, NE Energy will be able to negotiate prices that make building such a power hungry facility viable in Connecticut. NE Energy has also committed to paying a 12.08 percent premium to the town on top of what it pays Dominion for power, along with other payments said to total more than $1 billion over the next 30 years. But after initially denying the sale of land to NE Edge back in January over a lack of information regarding the datacenter project, it's reported that the town council has yet to tell the company what information it is after.

In a state where housing is expensive to build, to rent, or to buy — and not especially energy efficient — can a big blue robot make a difference?

The Boston Globe reports on Reframe Systems, one of the companies "trying robots to make construction more efficient" — in this case, "working alongside humans in an assembly line to build small houses in a factory." [Its cofounders] learned to get robots and humans to work together while at Amazon, which has built more than 750,000 bots in Massachusetts and deployed them to distribution centers around the world. Advising the company are Amy Villeneuve, former chief operating officer of that Amazon division, and Charly Mwangi, a veteran of the carmakers Nissan, Tesla, and Rivian...

Standing at one end of Reframe's factory, [cofounder Aaron] Small explained that the company's ambition is to build net-zero houses — houses that produce as much energy as they use — "twice as fast as traditional methods, twice as cheap, and with 10 times lower carbon" emissions. That means using large screws called helical piles to fix the house to the site, instead of a concrete foundation. (Concrete production generates large amounts of carbon dioxide.) The company buys recycled cellulose insulation to fill the walls. Solar panels go on the roof and triple-paned windows in the walls...

Reframe's "microfactory" can produce between 30 and 50 homes a year, [cofunder Vikas] Enti said. Eventually, the company aims to set up larger factories around the country, all within an hour's drive of big cities.
After a home is trucked to its final destination, "Electrical wires and plumbing are installed in both floors and walls as they're built," according to the article.

"Employees toting iPads can refer to digital construction drawings and get step-by-step instructions about tasks from cutting lumber to connecting pipes." One of the co-founders says, "We like to compare it to Lego instructions."

The Los Angeles Times checks in on America's largest dam-removal project, which they say is now "revealing a stark landscape that had been underwater for generations."

"A thick layer of muddy sediment covers the sloping ground, where workers have been scattering seeds and leaving meandering trails of footprints. In the cracked mud, seeds are sprouting and tiny green shoots are appearing." With water passing freely through tunnels in three dams, the Klamath River has returned to its ancient channel and is flowing unhindered for the first time in more than a century through miles of waterlogged lands. Using explosives and machinery, crews began blasting and tearing into the concrete of one of the three dams earlier this month... The emptying of the reservoirs, which began in January, is estimated to have released as much as 2.3 million tons of sediment into the river, abruptly worsening its water quality and killing nonnative perch, bluegill and bass that had been introduced in the reservoirs for fishing. Downstream from the dams, the river's banks are littered with dead fish. But tribal leaders, biologists and environmentalists say that this was part of the plan, and that the river will soon be hospitable for salmon to once again swim upstream to spawn... [The dams] blocked salmon from reaching vital habitat and degraded the river's water quality, contributing to toxic algae blooms in the reservoirs and disease outbreaks that killed fish...

Workers have been drilling holes in the top of the Copco No. 1 Dam, placing dynamite and setting off blasts, then using machinery to chip away fractured concrete. The dam, which has been in place since 1918, is scheduled to be fully removed by the end of August. The smaller Copco No. 2 Dam was torn down last year as the project began. Two earthen dams, the Iron Gate and the John C. Boyle, remain to be dismantled starting in May. If the project goes as planned, the three dams will be gone sometime this fall, reestablishing a free-flowing stretch of river and enabling Chinook and coho salmon to swim upstream and spawn along about 400 miles of the Klamath and its tributaries. Meanwhile, teams of scientists and workers are focusing on restoring the landscape and natural vegetation on about 2,200 acres of denuded reservoir-bottom lands...

River restoration advocates are optimistic. They say undamming the Klamath will demonstrate the potential for restoring free-flowing rivers elsewhere in California, and point to initial plans to remove two dams on the Eel River as another promising opportunity.

An anonymous reader quotes a report from InterestingEngineering: A concept that began as a doodle at a conference years ago is now becoming a reality. RocketStar Inc. has showcased (PDF) its advanced nuclear-based propulsion technology called the FireStar Drive. It is said to be the world's first electric device for spacecraft propulsion boosted by nuclear fusion. Recently, the company announced the successful initial demonstration of this electric propulsion technology.

The FireStar Drive harnesses the power of nuclear fusion to improve the performance of RocketStar's "water-fueled pulsed plasma thruster." A spacecraft's thrusters perform various functions, including propulsion, orbital changes, and even docking with other orbiting platforms. Moreover, the device employs a unique sort of aneutronic nuclear fusion, which is a fusion reaction that generates few to no neutrons as a byproduct. "The base thruster generates high-speed protons through the ionization of water vapor," noted the press release. Therefore, these protons collide with the nucleus of a boron atom, which starts the fusion reaction. The FireStar Drive begins a fusion process by adding boron into the thruster exhaust, resulting in high-energy particles that increase thrust.

RocketStar's current thruster is dubbed M1.5. Plans to test the FireStar Drive are now ongoing. The in-space technological demonstration will take place aboard D-Orbit's patented OTV ION Satellite Carrier. The SpaceX Transporter rideshare mission will likely launch the demo test in July and October 2024. Furthermore, the team plans to undertake ground tests this year, with more in-space demonstrations scheduled for February 2025. The FireStar Drive will undergo testing as a payload aboard Rogue Space System's Barry-2 spacecraft in the same month. The thruster M1.5 is already ready for delivery to clients.

After securing billions in federal grants and loans, Reuters reports that the company is "planning a $100-billion spending spree across four U.S. states" to build and expand its chip manufacturing factories. From the report: The centerpiece of Intel's five-year spending plan is turning empty fields near Columbus, Ohio, into what CEO Pat Gelsinger described to reporters on Tuesday as "the largest AI chip manufacturing site in the world," starting as soon as 2027. Intel's plan will also involve revamping sites in New Mexico and Oregon and expanding operations in Arizona, where longtime rival Taiwan Semiconductor Manufacturing Co is also building a massive factory that it hopes will receive funding from President Joe Biden's push to bring advanced semiconductor manufacturing back to the United States. [...]

Gelsinger said about 30% of the $100-billion plan will be spent on construction costs such as labor, piping and concrete. The remaining will go towards buying chipmaking tools from firms such as ASML, Tokyo Electron, Applied Materials and KLA, among others. Those tools will help bring the Ohio site online by 2027 or 2028, though Gelsinger warned the timeline could slip if the chip market takes a dive. Beyond grants and loans, Intel plans to make most of the purchases from its existing cash flows.

"It will still take three to five years for Intel to become a serious player in the foundry market" for cutting-edge chips, said Kinngai Chan, an analyst at Summit Insights. However, he warned more investment would be needed before Intel could overtake TSMC, adding that the Taiwanese firm could remain the leader for "some time to come." Gelsinger has previously said a second round of U.S. funding for chip factories would likely be needed to re-establish the U.S. as a leader in semiconductor manufacturing, which he reiterated on Tuesday. "It took us three-plus decades to lose this industry. It's not going to come back in three to five years of CHIPS Act" funding, said Gelsinger, who referred to the low-interest-rate funding as "smart capital."

Sean Hollister reports via The Verge: Nvidia's must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead -- with the new Blackwell B200 GPU and GB200 "superchip." Nvidia says the new B200 GPU offers up to 20 petaflops of FP4 horsepower from its 208 billion transistors and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It "reduces cost and energy consumption by up to 25x" over an H100, says Nvidia.

Training a 1.8 trillion parameter model would have previously taken 8,000 Hopper GPUs and 15 megawatts of power, Nvidia claims. Today, Nvidia's CEO says 2,000 Blackwell GPUs can do it while consuming just four megawatts. On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest seven times the performance of an H100, and Nvidia says it offers 4x the training speed. Nvidia told journalists one of the key improvements is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus, the 20 petaflops of FP4 I mentioned earlier). A second key difference only comes when you link up huge numbers of these GPUs: a next-gen NVLink switch that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth. That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia. Further reading: Nvidia in Talks To Acquire AI Infrastructure Platform Run:ai

Pi was calculated to 100 trillion decimal places in 2022 by a Google team lead by cloud developer advocate Emma Haruka Iwao.

But 2024's "pi day" saw a new announcement... After successfully breaking the speed record for calculating pi to 100 trillion digits last year, the team at StorageReview has taken it up a notch, revealing all the numbers of Pi up to 105 trillion digits! Spoiler: the 105 trillionth digit of Pi is 6!

Owner and Editor-in-Chief Brian Beeler led the team that used 36 Solidigm SSDs (nearly a petabyte) for their unprecedented capacity and reliability required to store the calculated digits of Pi. Although there is no practical application for this many digits, the exercise underscores the astounding capabilities of modern hardware and an achievement in computational and storage technology...

For an undertaking of this size, which took 75 days, the role of storage cannot be understated. "For the Pi computation, we're entirely restricted by storage, says Beeler. "Faster CPUs will help accelerate the math, but the limiting factor to many new world records is the amount of local storage in the box. For this run, we're again leveraging Solidigm D5-P5316 30.72TB SSDs to help us get a little over 1P flash in the system.

"These SSDs are the only reason we could break through the prior records and hit 105 trillion Pi digits."
"Leveraging a combination of open-source and proprietary software, the team at StorageReview optimized the algorithmic process to fully exploit the hardware's capabilities, reducing computational time and enhancing efficiency," Beeler says in the announcement.

There's a video on YouTube where the team discusses their effort.

The Washington Post reports: Nearly a decade ago, Congress passed a law that allows private American space companies the rights to resources they mine on celestial bodies, including the moon.

Now, there's a private venture that says it intends to do just that.

Founded by a pair of former executives from Blue Origin, the space venture founded by Jeff Bezos, and an Apollo astronaut, the company, Interlune, announced itself publicly Wednesday by saying it has raised $18 million and is developing the technology to harvest and bring materials back from the moon... Specifically, Interlune is focused on Helium-3, a stable isotope that is scarce on Earth but plentiful on the moon and could be used as fuel in nuclear fusion reactors as well as helping power the quantum computing industry.

The company, based in Seattle, has been working for about four years on the technology, which comes as the commercial sector is working with NASA on its goal of building an enduring presence on and around the moon... Rob Meyerson, the former president of Blue Origin, co-founded Interlune with Gary Lai, another former executive at Blue, and Harrison Schmitt, a geologist who flew to the moon during Apollo 17... In an interview, Meyerson said that the company intends to be the first to collect, return and then sell lunar resources and test the 2015 law. There is a large demand for Helium-3 in the quantum computing industry, which requires some of its systems to operate in extremely cold temperatures, and Interlune has already lined up a "customer that wants to buy lunar resources in large quantities," he said.

"We intend to be the first to go commercialize and deliver and support those customers," he said. NASA might want to be a customer as well. In 2020, it said it was looking for companies to collect rocks and dirt from the lunar surface and sell them to NASA as part of a technology development program that would eventually help astronauts "live off the land...." The company's funding round was led by the venture capital firm Seven Seven Six, whose founder and general partner, Alexis Ohanian, said that the space sector has become far more appealing to investors. "The space economy is something we can actually talk about with a straight face now, and I think some of the smartest people on the planet are making those efforts," he said...

He said he was aware that it might take years, or longer for a moon mining business to make money. But he said that, "we're comfortable waiting for a decade plus to see those returns."
NASA is planning more missions like the Intuitive Machines landing earlier this year, according to the article, "which it says will not only help pave the way for humans to return to the moon but for private industry to begin commercial operations there as well." Interlune plans "a prospecting machine" as soon as 2026, followed by an "end-to-end demonstration" in 2028 that harvests and returns a small quanity of Helium-3, and then full-scale operations by 2030.

"China has also said that it is interested in extracting other resources, including Helium-3, which it said was present in a sample it returned from the moon in 2020."

An anonymous reader quotes a report from NPR: America's first commercial-scale offshore wind farm is officially open, a long-awaited moment that helps pave the way for a succession of large wind farms. Danish wind energy developer Orsted and the utility Eversource built a 12-turbine wind farm called South Fork Wind 35 miles (56 kilometers) east of Montauk Point, New York. New York Gov. Kathy Hochul went to Long Island Thursday to announce that the turbines are delivering clean power to the local electric grid, flipping a massive light switch to "turn on the future." Interior Secretary Deb Haaland was also on hand.

Achieving commercial scale is a turning point for the industry, but what's next? Experts say the nation needs a major buildout of this type of clean electricity to address climate change. Offshore wind is central to both national and state plans to transition to a carbon-free electricity system. The Biden administration has approved six commercial-scale offshore wind energy projects, and auctioned lease areas for offshore wind for the first time off the Pacific and Gulf of Mexico coasts. New York picked two more projects last month to power more than 1 million homes. This is just the beginning, Hochul said. She said the completion of South Fork shows that New York will aggressively pursue climate change solutions to save future generations from a world that otherwise could be dangerous. South Fork can generate 132 megawatts of offshore wind energy to power more than 70,000 homes.

"It's great to be first, we want to make sure we're not the last. That's why we're showing other states how it can be done, why we're moving forward, on to other projects," Hochul told The Associated Press in an exclusive interview before the announcement. "This is the date and the time that people will look back in the history of our nation and say, 'This is when it changed,'" Hochul added. South Fork will generate more than four times the power of a five-turbine pilot project developed earlier off the coast of Rhode Island, and unlike that subsidized test project, was developed after Orsted and Eversource were chosen in a competitive bidding process to supply power to Long Island. Orsted CEO Mads Nipper called the opening a major milestone that proves large offshore wind farms can be built, both in the United States and in other countries with little or no offshore wind energy currently.
Another large U.S. offshore wind farm began producing energy in January, with plans to eventually power 62 turbines, enough to generate electricity for 400,000 homes.

Mercedes-Benz is the latest automotive company to trial how humanoid robots could be used to automate "low skill, physically challenging, manual labor." From a report: On Friday, robotics company Apptronik announced it had entered into a commercial agreement with Mercedes to pilot how "highly advanced robotics" like Apollo -- Apptronik's 160-pound bipedal robot -- can be used in manufacturing. The news follows a similar pilot announced by BMW in January.

Apptronik says that Mercedes is exploring use cases like having Apollo inspect and deliver components to human production line workers. Neither company has disclosed any figures for the agreement or how many Apollo robots are being trialed. According to Apptronik, humanoid robots would allow vehicle manufacturers to start automating manufacturing tasks without having to redesign their existing facilities. The company says its approach instead "centers on automating some physically demanding, repetitive and dull tasks for which it is increasingly hard to find reliable workers."

Dan Robinson reports via The Register: Adding caffeine can enhance the efficiency of fuel cells, reducing the need for platinum in electrodes and significantly reducing the cost of making them, according to researchers in Japan. [...] The study, published in the journal Communications Chemistry, concerns the catalysis process at the cathode of a fuel cell and making this reaction more efficient. Fuel cells work somewhat like batteries. They generate power by converting the chemical energy of a fuel (or electrolyte) and an oxidizing agent into electricity. This is typically hydrogen as a fuel and oxygen as an oxidizer. Unlike batteries with limited lifespans, fuel cells can generate power as long as fuel is supplied. The hydrogen undergoes oxidation at the anode, producing hydrogen ions and electrons. The ions move through the hydrogen electrolyte to the cathode, while the electrons flow through an external circuit, generating electricity. At the cathode, oxygen combines with the hydrogen ions and electrons, resulting in water as a by-product. However, this water impacts the performance of the fuel cell, reacting with the platinum (Pt) to form a layer of platinum hydroxide (PtOH) on the electrode and interfering with the catalysis of the oxygen reduction reaction (ORR), according to the researchers.

To maintain efficient operation, fuel cells require a high Pt loading (greater platinum content), which significantly ups the costs of fuel cells. A quick look online found market prices for platinum of $29.98 per gram, or $932.61 per ounce, at the time of writing. The researchers found that adding caffeine can improve the ORR activity of platinum electrodes 11 fold, making the reaction more efficient. If you are wondering (as we were) how they came to be experimenting with this, the paper explains that modifying electrodes with hydrophobic material is known to be an effective method for enhancing ORR. Caffeine is less toxic than other hydrophobic substances, and it activates the hydrogen evolution and oxidation reactions of Pt nanoparticles and caffeine doped carbons. Got that?

Chiba University's work was led by Professor Nagahiro Hoshi at the Department of Applied Chemistry and Biotechnology. He explained that the researchers found a notable improvement in the electrode's ORR activity with an increase in caffeine concentration in the electrolyte. This forms a thin layer on the electrode's surface, effectively preventing the formation of PtOH, but the effect depends on the orientation of the platinum atoms on the electrode's surface. The paper refers to these as Pt(100), Pt(110) and Pt(111), with the latter two showing increased ORR activity, while there was no noticeable effect with Pt(100). The researchers do not explain if this latter effect might be a problem, but instead claim that their discovery has the potential to improve the designs of fuel cells and lead to more widespread adoption.

The Federal Trade Commission and the Department of Justice have urged the US Copyright Office to broaden exemptions to the Digital Millennium Copyright Act's Section 1201. Specifically, the two agencies are advocating for the extension of the right to repair to include "commercial and industrial equipment," which includes McDonald's ice cream machines that are notorious for breaking down. The Verge reports: Exemptions to DMCA Section 1201 are issued every three years, as per the Register of Copyrights' recommendation. Prior exemptions have been issued for jailbreaking cellphones and repairing certain parts of video game consoles. The FTC and DOJ are asking the Copyright Office to go a step further, extending the right to repair to "commercial and industrial equipment." The comment (PDF) singles out four distinct categories that would benefit from DMCA exemptions: commercial soft serve machines; proprietary diagnostic kits; programmable logic controllers; and enterprise IT. 'In the Agencies' view, renewing and expanding repair-related exemptions would promote competition in markets for replacement parts, repair, and maintenance services, as well as facilitate competition in markets for repairable products," the comment reads.

The inability to do third-party repairs on these products not only limits competition, the agencies say, but also makes repairs more costly and can lead to hundreds or thousands of dollars in lost sales. Certain logic controllers have to be discarded and replaced if they break or if the passwords for them get lost. The average estimated cost of "unplanned manufacturing downtime" was $260,000 per hour, the comment notes, citing research from Public Knowledge and iFixit. As for soft serve machines, breakdowns can lead to $625 in lost sales each day. Business owners can't legally fix them on their own or hire an independent technician to do so, meaning they have to wait around for an authorized technician -- which, the comment says, usually takes around 90 days.

An anonymous reader quotes a report from Wired: We've been ableto design and 3D-print plastic phone cases and toys at home for a decade now. For almost every other consumer product made in a factory, the robots have taken over the heavy lifting. But fashion is still stuck in the 20th century. Take a typical pair of chinos. Cotton threads are woven on a large loom at a mill somewhere in Asia, then shipped to a dye house, then shipped (usually a great distance) to a garment factory somewhere else in Asia. There, the fabric is laid flat and cut into shapes, with the excess fabric being landfilled, incinerated, or (very rarely) recycled. Underpaid and exploited garment workers hand-sew those pieces of fabric into pants, which are then shipped across the ocean to a fulfillment warehouse or a store near you. This global apparel supply chain is inefficient and emissions-heavy -- an estimated 4 percent of global waste and 2 to 4 percent of greenhouse gas emissions are attributed to fashion production. Brands have to make risky predictions many months in advance about which items will sell, leading them to over order on a massive scale.

Now, Walmart is piloting a project with the San Francisco Bay area startup Unspun to test whether it can manufacture the retailer's in-house brand of chinos in the US using a technology called 3D weaving. The experiment is part of a push to nearshore Walmart's supply chain and cut down on emissions and waste associated with textile production. While still very much in the prototype phase -- the two companies are exploring how to use Unspun's technology to supply pants to Walmart's stores -- if successful, this project could upend the way apparel is manufactured on a huge scale. Unspun hopes to eventually deploy 3D weaving micro-factories throughout the United States, so that anyone can order custom and locally made apparel on demand.

According to Bloomberg (paywalled), the Pentagon has reportedly scrapped its plan to allocate $2.5 billion in grants to Intel, causing the firm's stock to slip in extended-hours trading. From a report: The decision now leaves the U.S. Commerce Department, which is responsible for doling out the funds from the U.S. CHIPs and Science Act, to make up the shortfall, the news outlet said. The Commerce Dept. was initially only supposed to cover $1B of the $3.5B that Intel is slated to receive for advanced defense and intelligence-related semiconductors. The deal is slated to position Intel as the dedicated supplier for processors used for military and intelligence applications and could result in a Secure Enclave inside Intel's chip factory, the news outlet said. With the Pentagon reportedly pulling out, it could alter how much Intel and other companies receive from the CHIPs Act, the news outlet said.

After 10 years and billions of dollars spent in development, Apple abruptly canceled its ambitious car project known as "Titan," shifting its focus and resources on the company's artificial intelligence division. In a recent Q&A on Monday, Bloomberg's Mark Gurman (paywalled) shared some new insights about the project and how involved the Apple Silicon team was before it was shut down. According to Gurman, Apple was planning to power the "AI brain" of the car with a custom Apple Silicon chip that would have the equivalent power of four M2 Ultra chips (the most powerful Apple has to date) combined. 9to5Mac reports: A single M2 Ultra chip consists of 134 billion transistors and features a 24-core CPU, a GPU with up to 76 cores, and a dedicated 32-core Neural Engine. M2 Ultra powers the current generation of Mac Studio and Mac Pro. Interestingly, Gurman says that the development of this new chip for the car was "nearly finished" before the project was discontinued. As some of the engineers working on the car project were reassigned to other teams at Apple, the company could reuse the engineering of this new chip for future projects.

Over 15,000 Roku customers were hacked and used to make fraudulent purchases of hardware and streaming subscriptions. According to BleepingComputer, the threat actors were "selling the stolen accounts for as little as $0.50 per account, allowing purchasers to use stored credit cards to make illegal purchases." From the report: On Friday, Roku first disclosed the data breach, warning that 15,363 customer accounts were hacked in a credential stuffing attack. A credential stuffing attack is when threat actors collect credentials exposed in data breaches and then attempt to use them to log in to other sites, in this case, Roku.com. The company says that once an account was breached, it allowed threat actors to change the information on the account, including passwords, email addresses, and shipping addresses. This effectively locked a user out of the account, allowing the threat actors to make purchases using stored credit card information without the legitimate account holder receiving order confirmation emails.

"It appears likely that the same username/password combinations had been used as login information for such third-party services as well as certain individual Roku accounts," reads the data breach notice. "As a result, unauthorized actors were able to obtain login information from third-party sources and then use it to access certain individual Roku accounts. "After gaining access, they then changed the Roku login information for the affected individual Roku accounts, and, in a limited number of cases, attempted to purchase streaming subscriptions." Roku says that it secured the impacted accounts and forced a password reset upon detecting the incident. Additionally, the platform's security team investigated for any charges due to unauthorized purchases performed by the hackers and took steps to cancel the relevant subscriptions and refund the account holders.

A researcher told BleepingComputer last week that the threat actors have been using a Roku config to perform credential stuffing attacks for months, bypassing brute force attack protections and captchas by using specific URLs and rotating through lists of proxy servers. Successfully hacked accounts are then sold on stolen account marketplaces for as little as 50 cents, as seen below where 439 accounts are being sold. The seller of these accounts provides information on how to change information on the account to make fraudulent purchases. Those who purchase the stolen accounts hijack them with their own information and use stored credit cards to purchase cameras, remotes, soundbars, light strips, and streaming boxes. After making their purchases, it is common for them to share screenshots of redacted order confirmation emails on Telegram channels associated with the stolen account marketplaces.

An anonymous reader shared this article from Futurism: More than two years since MIT claimed its scientists achieved a breakthrough in fusion energy, the university is claiming that new research "confirms" that the magnet-based design used in those tests isn't just impressive in a lab setting, but is practical and economically viable, too.

These findings come from a comprehensive report which features six separate [peer-reviewed] studies published in the journal IEEE Transactions on Applied Superconductivity this month, assessing the feasibility of the superconductor magnets used by MIT scientists in their landmark test conducted in September 2021.
"Together, the papers describe the design and fabrication of the magnet and the diagnostic equipment needed to evaluate its performance," MIT announced, "as well as the lessons learned from the process.

"Overall, the team found, the predictions and computer modeling were spot-on, verifying that the magnet's unique design elements could serve as the foundation for a fusion power plant." The successful test of the magnet, says Hitachi America Professor of Engineering Dennis Whyte, who recently stepped down as director of MIT's Plasma Science and Fusion Center, was "the most important thing, in my opinion, in the last 30 years of fusion research." Before the [2021] demonstration, the best-available superconducting magnets were powerful enough to potentially achieve fusion energy — but only at sizes and costs that could never be practical or economically viable. Then, when the tests showed the practicality of such a strong magnet at a greatly reduced size, "overnight, it basically changed the cost per watt of a fusion reactor by a factor of almost 40 in one day," Whyte says.

"Now fusion has a chance," Whyte adds

A recent survey found that academic organizations are failing to preserve digital material -- "including science paid for with taxpayer money," reports Ars Technica, highlighting the need for improved archiving standards and responsibilities in the digital age. From the report: The work was done by Martin Eve, a developer at Crossref. That's the organization that organizes the DOI system, which provides a permanent pointer toward digital documents, including almost every scientific publication. If updates are done properly, a DOI will always resolve to a document, even if that document gets shifted to a new URL. But it also has a way of handling documents disappearing from their expected location, as might happen if a publisher went bankrupt. There are a set of what's called "dark archives" that the public doesn't have access to, but should contain copies of anything that's had a DOI assigned. If anything goes wrong with a DOI, it should trigger the dark archives to open access, and the DOI updated to point to the copy in the dark archive. For that to work, however, copies of everything published have to be in the archives. So Eve decided to check whether that's the case.

Using the Crossref database, Eve got a list of over 7 million DOIs and then checked whether the documents could be found in archives. He included well-known ones, like the Internet Archive at archive.org, as well as some dedicated to academic works, like LOCKSS (Lots of Copies Keeps Stuff Safe) and CLOCKSS (Controlled Lots of Copies Keeps Stuff Safe). The results were... not great. When Eve broke down the results by publisher, less than 1 percent of the 204 publishers had put the majority of their content into multiple archives. (The cutoff was 75 percent of their content in three or more archives.) Fewer than 10 percent had put more than half their content in at least two archives. And a full third seemed to be doing no organized archiving at all. At the individual publication level, under 60 percent were present in at least one archive, and over a quarter didn't appear to be in any of the archives at all. (Another 14 percent were published too recently to have been archived or had incomplete records.)

The good news is that large academic publishers appear to be reasonably good about getting things into archives; most of the unarchived issues stem from smaller publishers. Eve acknowledges that the study has limits, primarily in that there may be additional archives he hasn't checked. There are some prominent dark archives that he didn't have access to, as well as things like Sci-hub, which violates copyright in order to make material from for-profit publishers available to the public. Finally, individual publishers may have their own archiving system in place that could keep publications from disappearing. The risk here is that, ultimately, we may lose access to some academic research.

An anonymous reader quotes a report from Ars Technica: AMD announced this week that it has ceased FreeSync certification for monitors or TVs whose maximum refresh rates are under 144 Hz. Previously, FreeSync monitors and TVs could have refresh rates as low as 60 Hz, allowing for screens with lower price tags and ones not targeted at serious gaming to carry the variable refresh-rate technology. AMD also boosted the refresh-rate requirements for its higher AdaptiveSync tiers, FreeSync Premium and FreeSync Premium Pro, from 120 Hz to 200 Hz.

Here are the new minimum refresh-rate requirements for FreeSync, which haven't changed for laptops. AMD will continue supporting already-certified FreeSync displays even if they don't meet the above requirements. Interestingly, AMD's minimum refresh-rate requirements for TVs goes beyond 120 Hz, which many premium TVs max out at currently, due to the current-generation Xbox and PlayStation supporting max refresh rates of 120 frames per second (FPS). Announcing the changes this week in a blog post, Oguzhan Andic, AMD FreeSync and Radeon product marketing manager, claimed that the changes were necessary, noting that 60 Hz is no longer "considered great for gaming." Andic wrote that the majority of gaming monitors are 144 Hz or higher, compared to in 2015, when FreeSync debuted, and even 120 Hz was "a rarity."

An anonymous reader quotes a report from The Week Magazine: The advancement of new technologies appears to have given rise to a new problem across the United States: a crippling power shortage on the horizon. The advent of these technologies, such as eco-friendly factories and data centers, has renewed concerns that America could run out of electrical power. These worries also come at a time when the United States' aging power grid is in desperate need of repair. Heavily publicized incidents such as the 2021 Texas power outage, which was partially blamed on crypto-farming, exposed how vulnerable the nation's power supply is, especially during emergencies. There have also been warnings from tech moguls such as Elon Musk, who has stated that the United States is primed to run out of electricity and transformers for artificial intelligence in 2025. But the push to extend the life of the nation's power grid, while also maintaining eco-friendly sustainability, begs the question: Is the United States really at risk of going dark?

The emergence of new technologies means demand is soaring for power across the country; in Georgia, "demand for industrial power is surging to record highs, with the projection of electricity use for the next decade now 17 times what it was only recently," Evan Halper said for The Washington Post. Northern Virginia "needs the equivalent of several large nuclear power plants to serve all [its] new data centers," Halper said, while Texas faces a similar problem. This demand is resulting in a "scramble to try to squeeze more juice out of an aging power grid." At the same time, companies are "pushing commercial customers to go to extraordinary lengths to lock down energy sources, such as building their own power plants," Halper said. Much of this relates to the "rapid innovation in artificial intelligence, which is driving the construction of large warehouses of computing infrastructure," Halper said. This infrastructure requires significantly more power than traditional data centers, with the aforementioned crypto farms also sucking up massive amounts of power.

Climate change is also hurting sustainability efforts. A recent report from the North American Electric Reliability Corporation estimated that more than 300 million people in the U.S. and Canada could face power shortages in 2024. It also found that electricity demand is rising faster now than at any time in the past five years. This is partially because the "push for the electrification of heating and transportation systems -- including electric cars -- is also creating new winter peaks in electricity demand," Jeremy Hsu said for New Scientist. One of the main issues with these sustainability efforts is the push to move away from fossil fuels toward renewable power. Natural gas is often seen as a bridge between fossils and renewables, but this has also had unintended consequences for the power grid. The system delivering natural gas "doesn't have to meet the same reliability standards as the electric grid, and in many cases, there's no real way to guarantee that fuel is available for the gas plants in the winter," Thomas Rutigliano of the Natural Resources Defense Council said to New Scientist. As a result, the "North American electricity supply has become practically inseparable from the natural gas supply chain," John Moura of the North American Electric Reliability Corporation said to New Scientist. As such, a "reliable electricity supply that lowers the risk of power outages depends on implementing reliability standards for the natural gas industry moving forward," but this may be easier said than done.