Slashdot

Sony Interactive Entertainment (basically the PlayStation wing) today announced plans to acquire Audeze. From a report: The Orange Country, CA-based firm is best known for its high-end headphones, targeted at gamers and audio producers. The hardware is often priced into four-figures, but the company earns consistently high marks from reviewers and users alike. In a release announcing the news, the companies are quick to note that operating under the Sony banner won't mean platform exclusivity (at least not at first). Audeze will continue to produce gaming headphones for competitors, including -- presumably -- the Xbox. Playstation branded products, however, are almost certain to follow.Read more of this story at Slashdot.

Gmail only asks for your user credentials during the initial login, and that login session can last for weeks at a time. That's not as secure as it could be, so soon Gmail will start posting 2FA challenges if you try to access any "sensitive" settings, even when you're already logged in. From a report: The newly protected settings are for filters, account forwarding, and IMAP. Soon, poking around in any of these options will boot you into a "Verify it's you" 2FA prompt, and you'll have to pass the challenge on your phone (these settings are only available on the web). If this 2FA challenge is failed or not answered, you'll get a bright red "Critical security alert" pop-up alerting you to the attempt on all your trusted devices.Read more of this story at Slashdot.

Zadie Smith, Stephen King, Rachel Cusk and Elena Ferrante are among thousands of authors whose pirated works have been used to train artificial intelligence tools, a story in The Atlantic has revealed. The Guardian: More than 170,000 titles were fed into models run by companies including Meta and Bloomberg, according to an analysis of "Books3" -- the dataset harnessed by the firms to build their AI tools. Books3 was used to train Meta's LLaMA, one of a number of large language models -- the best-known of which is OpenAI's ChatGPT -- that can generate content based on patterns identified in sample texts. The dataset was also used to train Bloomberg's BloombergGPT, EleutherAI's GPT-J and it is "likely" it has been used in other AI models. The titles contained in Books3 are roughly one-third fiction and two-thirds nonfiction, and the majority were published within the last two decades. Along with Smith, King, Cusk and Ferrante's writing, copyrighted works in the dataset include 33 books by Margaret Atwood, at least nine by Haruki Murakami, nine by bell hooks, seven by Jonathan Franzen, five by Jennifer Egan and five by David Grann. Books by George Saunders, Junot DAaz, Michael Pollan, Rebecca Solnit and Jon Krakauer also feature, as well as 102 pulp novels by Scientology founder L Ron Hubbard and 90 books by pastor John MacArthur. The titles span large and small publishers including more than 30,000 published by Penguin Random House, 14,000 by HarperCollins, 7,000 by Macmillan, 1,800 by Oxford University Press and 600 by Verso.Read more of this story at Slashdot.

Mastercard and crypto exchange Binance will end their four crypto card programmes in Argentina, Brazil, Colombia and Bahrain as of Sept. 22. From a report: The Binance cards allow users to make payments in traditional currencies, funded by their cryptocurrency holdings on the exchange. Mastercard's website also lists partnerships with crypto exchanges including Gemini. The decision will not impact any of Mastercard's other crypto card programmes, the spokesperson said. Binance is facing legal and regulatory challenges.Read more of this story at Slashdot.

Shares of Nvidia rose 6.7% to hit a new all-time high on Thursday, after the company unveiled a $25 billion buyback plan, coupled with record quarterly revenue powered by strong demand for its artificial intelligence (AI)-focused chips. From a report: Nvidia generated revenue of $13.51 billion in the second quarter and predicted that revenue would reach $16 billion in the third quarter, the company reported late on Wednesday. Both results exceeded analyst expectations, as per Refinitiv. Santa-Clara, California-based Nvidia also said it would buyback $25 billion worth of its shares. Nvidia's stock rose as high as $502.66, surpassing a record high hit earlier this week, and further entrenching itself as the first trillion-dollar chip maker. It pared some of those gains and was last up 1.8% at $479.82. The stock is now up 227% year-to-date.Read more of this story at Slashdot.

An anonymous reader shares a report: German goals to cut greenhouse emissions by 65% by 2030 are likely to be missed, meaning a longer-term net zero by a 2045 target is also in doubt, reports by government climate advisers and the Federal Environment Agency (UBA) show. The European Union has sought to be a climate leader and Germany has set itself more ambitious targets than the bloc as a whole, but in many countries politics and the economic crisis have pushed the climate crisis down the agenda. Germany, Europe's largest economy, aims to cut its carbon dioxide emissions by 65% by 2030 compared with 1990. Last year its CO2 levels were already 40% below the 1990 level, but the new reports said that was not enough. "The expected overall reduction is probably overestimated," Hans-Martin Henning, the chairman of a council of climate experts that advises the government, said in a statement on Tuesday. The German government has ordered 130 measures in various sectors. The buildings and transport sectors in particular are failing to implement them, the council of government climate advisers' report said. The buildings sector is expected to be 35 million tonnes of CO2 short of target by 2030, while the transport sector is expected to have excess emissions of between 117 million and 191 million tonnes compared with the government target. Tuesday's advisers' report coincided with another from the UBA that found that Germany cannot become climate neutral by 2045 on the basis of planned and existing government climate policy.Read more of this story at Slashdot.

Dropbox, a provider of online data storage, is ending its unlimited option, saying a small handful of customers were using massive amounts of resources that had the potential to degrade the cloud service for the rest of its clients. From a report: The company's highest-tier "all the space you need" storage plan will be capped at about 5 terabytes per user for new customers, the company said in a blog post. While the plan was designed for businesses, some clients were instead using it for cryptocurrency mining, pooling storage with strangers, or re-selling the cloud service, Dropbox said. These uses "frequently consume thousands of times more storage than our genuine business customers, which risks creating an unreliable experience for all of our customers," the company said. [...] The change follows Alphabet's Google removing "as much storage as you need" product branding for its highest-tier Workspace plan in May, according to copies of its website hosted on the Wayback Machine.Read more of this story at Slashdot.

An anonymous reader shares a report: For a decade until 2018, China sought to recruit elite foreign-trained scientists under a lavishly funded program that Washington viewed as a threat to U.S. interests and technological supremacy. Two years after it stopped promoting the Thousand Talents Plan (TTP) amid U.S. investigations of scientists, China quietly revived the initiative under a new name and format as part of a broader mission to accelerate its tech proficiency, according to three sources with knowledge of the matter and a Reuters review of over 500 government documents spanning 2019 to 2023. The revamped recruitment drive, reported in detail by Reuters for the first time, offers perks including home-purchase subsidies and typical signing bonuses of 3 to 5 million yuan, or $420,000 to $700,000, the three people told Reuters. China operates talent programs at various levels of government, targeting a mix of overseas Chinese and foreign experts. The primary replacement for TTP is a program called Qiming overseen by the Ministry of Industry and Information Technology, according to national and local policy documents, online recruitment advertisements and a person with direct knowledge of the matter who, as with others, spoke on the condition of anonymity because of the issue's sensitivity. The race to attract tech talent comes as President Xi Jinping emphasises China's need to achieve self-reliance in semiconductors in the face of U.S. export curbs. Regulations adopted by the U.S. Commerce Department in October restrict U.S. citizens and permanent residents from supporting the development and production of advanced chips in China, among other measures.Read more of this story at Slashdot.

Fiber-optic cables that were damaged by a rockfall in an undersea canyon, resulting in slow internet connections in some parts of Africa, should be repaired next month by a specialized vessel, according to telecommunication companies. From a report: The West Africa Cable System that runs about 16,000 kilometers (9,950 miles) along the sea floor from Europe to southern Africa was damaged with other lines earlier this month. The 40-year-old cable-layer vessel Leon Thevenin, named after a French telegraph engineer, was moored in Cape Town this week, according to tracking data compiled by Bloomberg. It's capable of working in extreme conditions and in shallow or deep water, according to owner Orange Marine, a submarine telecommunications company. All South African networks are currently experiencing disruptions due to the damaged lines, said Anne-Caroline Tanguy, a spokeswoman at Cloudflare, a company that provides load balancing and analysis. The repairs are expected to be finished in September.Read more of this story at Slashdot.

Meta, intent on making a splash in a generative AI space rife with competition, is on something of an open source tear. From a report: Following the release of AI models for generating text, translating languages and creating audio, the company today open sourced Code Llama, a machine learning system that can generate and explain code in natural language -- specifically English. Akin to GitHub Copilot and Amazon CodeWhisperer, as well as open source AI-powered code generators like StarCoder, StableCode and PolyCoder, Code Llama can complete code and debug existing code across a range of programming languages, including Python, C++, Java, PHP, Typescript, C# and Bash. "At Meta, we believe that AI models, but large language models for coding in particular, benefit most from an open approach, both in terms of innovation and safety," Meta wrote in a blog post shared with TechCrunch. "Publicly available, code-specific models can facilitate the development of new technologies that improve peoples' lives. By releasing code models like Code Llama, the entire community can evaluate their capabilities, identify issues and fix vulnerabilities." Code Llama, which is available in several flavors, including a version optimized for Python and a version fine-tuned to understand instructions (e.g. "Write me a function that outputs the fibonacci sequence"), is based on the Llama 2 text-generating model that Meta open sourced earlier this month. While Llama 2 could generate code, it wasn't necessarily good code -- certainly not up to the quality a purpose-built model like Copilot could produce.Read more of this story at Slashdot.

T-Mobile USA is cutting 7% of its staff, part of an effort to rein in costs as the company spends heavily to attract new subscribers in an increasingly competitive market. From a report: The move will affect about 5,000 positions, mostly corporate and back-office staff along with some technology roles, T-Mobile said in a regulatory filing. Retail and consumer-care experts won't be impacted. Chief Executive Officer Mike Sievert said in a letter to employees that the cost of attracting and retaining customers is "materially more expensive than it was just a few quarters ago." Building out the company's high-speed internet business and efforts in other areas "is not enough to deliver on these changing customer expectations going forward," he said.Read more of this story at Slashdot.

An anonymous reader shares a report: MSI made a mistake of sharing an unlisted video which has now leaked out. This leak involves a product training video showing MSI's latest Intel 700 motherboard series and upcoming PC cases. While the majority of the video focuses on improvements for these motherboards, there is a slide explaining Intel's next-gen Core series in a very short form. A slide (check the linked news post) has confirmed that Intel's 14th Gen Core series will see no major core count upgrades. Only one of the upcoming K-series CPUs will see a core change for its hybrid configuration of Performance and Efficient. The Core i7-14700K is getting 12 Efficient cores, which is an upgrade over 8 E-cores the Core i7-13700K. The Core i9-14900K, now confirmed by MSI as well, will use the same 8P+16E configuration. This also applies to Core i5-14600K with 6P+8E config. [...] The CPU series will use the same Intel 7 process technology and will only provide higher DDR5 frequency support. The company confirms that Intel 14th Gen Core is 3% faster on average compared to current-gen series. The most important upgrade involves the Core i7-14700K which has up to 17% faster multi-threaded (MT) performance due to extra cores.Read more of this story at Slashdot.

samleecole shares a report from 404 Media, a new independent media company founded by technology journalists Jason Koebler, Emanuel Maiberg, Samantha Cole, and Joseph Cox: On the Discord server for Mage Space, a popular platform for creating AI-generated images, is a list of channels where members share adult content. There are channels for furries, hardcore and softcore porn, and anime. At the bottom of the list is a channel named "other-nsfw" which includes a few distinct erotic genres that don't fit neatly into any of the others. Mostly, it's gore, violence, and bizarre, in hyperrealistic erotic imagery entirely generated by AI. The images people create, and the long, meandering prompts they write, are a rich text that could offer a glimpse into where sexuality in the internet age is taking us next, and how we're steering it. There's no shortage of fetish content on the internet, which might make the above statement sound ridiculous and unbelievable. Online, fetishists find their people and set to work making more of what they like, whether it's elaborate role-playing cosplays of themselves as sexy airplanes, blueberries, or slime monsters. Sometimes it pushes the bounds of the sane and legal: crush, fart, and scat porn all thrive online, and snuff films have been popular since before the internet existed. But with the rise in popularity of generative AI, and wildly popular sites like Mage.Space that let users generate anything they set their minds to, the limits are literally our own imaginations. With that power, people are wrangling images out of the AI that are on the edge of what's popular, let alone possible in the porn world. "This conversation we're having is literally on the edge of knowledge, nobody's writing about this in academia right now," Thomas Brooks, assistant professor of psychology at New Mexico Highlands University, told me. "Everybody's still kind of caught up in deepfakes. And they haven't quite grappled with generative AI yet." "You, as the individual porn consumer, can now create your own special little fantasy and your own technological, disembodied sexuality," said Brooks, in what he refers to as gamified pornography. "There's an internal motivation to solve the puzzle and get the prize. But then there's an external motivation of, 'can I come up with this crazy thing to show my anonymous internet friends.'" "We're letting technology become mediators of our different psychosocial expressions," Brooks added.Read more of this story at Slashdot.

According to The Information (paywalled), SpaceX is working with Cloudlfare to boost the performance of its satellite internet service Starlink. Reuters reports: The two companies are working on a way to increase Starlink's network of mini data centers around the globe that could help it deliver faster network speeds to its customers, the report said. According to SpaceX's website, Starlink users typically have download speeds between 25 and 220 Mbps, with the "majority" over 100 Mbps. Upload speeds range between 5 and 20 Mbps.Read more of this story at Slashdot.

An anonymous reader quotes a report from The Guardian: A severely paralyzed woman has been able to speak through an avatar using technology that translated her brain signals into speech and facial expressions. The latest technology uses tiny electrodes implanted on the surface of the brain to detect electrical activity in the part of the brain that controls speech and face movements. These signals are translated directly into a digital avatar's speech and facial expressions including smiling, frowning or surprise. The patient, a 47-year-old woman, Ann, has been severely paralyzed since suffering a brainstem stroke more than 18 years ago. She cannot speak or type and normally communicates using movement-tracking technology that allows her to slowly select letters at up to 14 words a minute. She hopes the avatar technology could enable her to work as a counsellor in future. The team implanted a paper-thin rectangle of 253 electrodes on to the surface of Ann's brain over a region critical for speech. The electrodes intercepted the brain signals that, if not for the stroke, would have controlled muscles in her tongue, jaw, larynx and face. After implantation, Ann worked with the team to train the system's AI algorithm to detect her unique brain signals for various speech sounds by repeating different phrases repeatedly. The computer learned 39 distinctive sounds and a Chat GPT-style language model was used to translate the signals into intelligible sentences. This was then used to control an avatar with a voice personalized to sound like Ann's voice before the injury, based on a recording of her speaking at her wedding. The technology was not perfect, decoding words incorrectly 28% of the time in a test run involving more than 500 phrases, and it generated brain-to-text at a rate of 78 words a minute, compared with the 110-150 words typically spoken in natural conversation. However, scientists said the latest advances in accuracy, speed and sophistication suggest the technology is now at a point of being practically useful for patients. A crucial next step is to create a wireless version of the BCI that could be implanted beneath the skull. The findings have been published in the journal Nature.Read more of this story at Slashdot.

Amazon has filed a lawsuit against a group of online stores that sell pirated DVDs of key titles such as "The Lord of the Rings: The Rings of Power" and "The Peripheral." TorrentFreak reports: In a complaint filed at a California federal court, Amazon accuses seven websites of selling pirated discs. These sites, including dvdshelf.com.au, dvds.trade, and dvdwholesale.co.uk, are presumably operated by the same group, using a variety of companies. For the public at large, it may not be immediately obvious that these discs are pirated. However, since Amazon doesn't produce or sell DVDs for these Prime Video series, there is no doubt that they are created from illicit sources. The piracy operation consists of at least seven websites and these all remain online today. According to Amazon, the sites ship to customers in the U.S. and abroad, twenty-four hours a day, seven days a week, resulting in mass copyright infringement. Before going to court, investigators conducted more than twenty test purchases of pirated DVDs. After these orders arrived, Amazon sent the discs to the Motion Picture Association which independently confirmed that they were all pirated. The complaint lists Yangchun Zhang as a key suspect. This person presumably resides in China and obtained the 'DVD Shelf' trademark in Australia. In addition, Zhang is also listed as the registrant of several of the domain names involved. The complaint accuses Zhang and the others of both copyright and trademark infringement. Through the lawsuit (PDF), Amazon hopes to recoup damages, which can run in the millions of dollars. Another key priority is to shut the sites down and Amazon asks the court for an injunction to stop all infringing activity.Read more of this story at Slashdot.

samleecole shares a report from 404 Media, a new independent media company founded by technology journalists Jason Koebler, Emanuel Maiberg, Samantha Cole, and Joseph Cox: Apple told a California legislator that it is formally supporting a right to repair bill in California, a landmark move that suggests big tech manufacturers understand they have lost the battle to monopolize repair, and need to allow consumers and independent repair shops to fix their own electronics. "Apple writes in support of SB 244, and urges members of the California legislature to pass the bill as currently drafted," Apple wrote to Susan Eggman, the sponsor of the bill, in a letter obtained by 404 Media. "We support SB 244 because it includes requirements that protect individual users' safety and security, as well as product manufacturers' intellectual property. We will continue to support the bill, so long as it continues to provide protections for customers and innovators." This is a landmark shift in policy from Apple, the most powerful electronics manufacturer in the world and, historically, one of the biggest opponents of right to repair legislation nationwide. It means, effectively, that consumers have won. "If California votes yes and continues to raise the bar on electronics repair from other states, it's becoming obvious the fight is over, and that we've won," said Nathan Proctor, Senior Director of consumer rights group U.S. PIRG Campaign for the Right to Repair. "It's going to be show over for consumer electronics. There are other industries where this fight is going to continue, but if a strong bill passes in California, we're winning." "I would think that passage in California means there'd be a lot of pressure on manufacturers to kind of set the line there and say 'no farther,' because we've now proven to them we can pass laws and change the ways they have to operate," Proctor added. "This shows state advocacy is a good way to deal with large problems that are hard to get through Congress. It shows you can really spread big tech thin if you have a real grassroots network behind you." iFixit and TechCrunch first reported the news.Read more of this story at Slashdot.

A new report from Business Insider (paywalled) describes how Google's employees were "frustrated" at Google's lack of progress when the Vision Pro was unveiled and provides a glimpse of what Google's current plans for an AR product are. Ars Technica reports: The BI report details how Google's latest dead project, Iris, "was beset by a constantly shifting strategy and lack of focus from senior leadership." After "conversations with seven current and former employees close to Google's AR efforts," Business Insider quotes a few of those anonymous employees, with one saying, "Every six months there was a major pivot in the program." At one point Google was working on a pair of custom silicon chips for the glasses' display and compute power and then gave up on the idea of custom chips. That work was apparently near completion, with one person saying, "I think it's weird when you convince yourselves you need to build custom silicon, and then you go and do that -- and then flush it down the toilet." Display problems led the team to switch from regular eyeglasses to sunglasses and then back again, and the team couldn't settle on a color or monochrome display. Google showed off a pair of Iris glasses at Google I/O that could translate spoken language, then quickly canned the idea. You might think Bavor leaving in February would be good, considering how little traction the AR division managed in the marketplace, but apparently the executive's departure created a "state of chaos" in the division. Google's next AR pivot is a partnership with Samsung, another company that has dabbled in AR/VR for years yet has no current product line. Google, Samsung, and Qualcomm have already vaguely announced an Apple-fighting mixed-reality partnership in February. Plans to actually launch a headset were reportedly delayed in the wake of the Vision Pro unveiling due to the headset not being competitive. The new launch target is sometime around summer 2024, but the report says that "some employees are skeptical [that] will be enough time to launch a product that will wow the public." According to the report, Samsung wants to follow its usual strategy and "build a headset device similar to Apple's Vision Pro." The project is apparently code-named "Moohan," and if you couldn't already guess from this lineup of companies, it will run Android. Despite acquiring hardware companies like the Micro-LED manufacturer Raxiom and smart glasses-maker North, Google now wants to "pivot to software" and follow the Android model. The partnership with Samsung makes Moohan the most likely project to actually hit the market, but Google still has two other competing XR projects. Raxiom also is apparently still around and works under Paul Greco, Magic Leap's former chief technology officer. Iris' software work has moved to "a new team" and is being turned into a software project codenamed "Betty" that Google wants to pitch to other manufacturers. Samsung doesn't want any of these other parts of Google or other hardware competitors to be privy to its Vision Pro clone, so the three teams are all firewalled off from each other and have to compete for resources. One current employee described the whole situation as "a weird bureaucratic mess."Read more of this story at Slashdot.

An anonymous reader quotes a report from TechCrunch: The National Highway Traffic Safety Administration (NHTSA) said Tuesday automakers can comply with a Massachusetts Right to Repair law, reversing a previous directive to ignore the state legislation. Massachusetts's Right to Repair law was a ballot initiative that passed overwhelmingly in 2020. The law requires auto manufacturers that sell cars in the state to equip vehicles with a standardized open data platform so that owners and independent mechanics can access telematics data for repairs, maintenance and diagnostics. In June 2023, NHTSA told automakers they needn't comply with the law, citing hacking concerns. The agency claimed sharing vehicle data would enable criminals to steal data or take control of cars remotely. NHTSA now says the law can roll out, with some caveats. Automakers can safely share diagnostic data with independent mechanics using short-range wireless technology. Long-range wireless signals, though, could potentially allow hackers to send dangerous commands to moving vehicles. The auto safety agency also said automakers should be allowed "a reasonable period of time" to put the technology in place. "[The U.S. Department of Transportation] strongly supports the right to repair and is eager to promote consumers' ability to choose independent or DIY repairs without compromising safety to themselves or others on our nation's roads," said Ben Halle, director of public affairs at USDOT. "The clarifications contained in the exchange of letters between state and federal partners ensure a path forward to promote competition and give consumers more options, while mitigating a dangerous risk to safety."Read more of this story at Slashdot.

Hollywood studios have presented a new proposal to writers that includes the highest wage increase in 35 years, protections against the impact of artificial intelligence, and other provisions. CBS News reports: Writers have been picketing outside major studios for over 100 days, surpassing the 2007-2008 strike. One of the major sticking points between the two sides was their stark differences in wage increases and residuals. The proposal sent to the Writers Guild of America on Aug. 11 includes a 5% increase in the first year of the contract, then 4% the next year, and 3.5% in the third, totaling a compounded 13% increase. Before the WGA went on strike on May 2, the AMPTP offered writers 4%-3%- 2% in the respective years, or 9% over the duration of the contract. The recent offer does not match the WGA's demand of 6%-5%-5% in the respective years but does bring them from $9,888 a week to $11,371 a week for guarantees of up to 9 weeks. They also moved to guarantee writers a minimum of 10 weeks of employment, a proposal they initially refused before the strike. AMPTP also increased the total domestic and foreign residuals for writers from $72,067 to $87,546 per episode over three years. Additionally, the union seemed to cave on the WGA's proposal to implement a viewership-based streaming residuals model. "For the first time, viewership data in the form of quarterly confidential reports is to be provided to the WGA that will include total SVOD view hours per title. This increased transparency will enable the WGA to develop proposals to restructure the current SVOD residual regime in the future," AMPTP wrote in the offer. Previously, the studios flat-out rejected the proposal and refused to make a counter, according to the WGA. Studios also included a tenet regarding artificial intelligence protections in the proposed deal. "The Companies confirm that because [Generative Artificial Intelligence] is not a person, it is not a 'writer' or 'professional writer' as defined in this MBA and, therefore, written material produced by GAI will not be considered literary material under this or any prior MBA," the AMPTP wrote in the offer. The union continued: "The proposal provides important safeguards to prevent writers from being disadvantaged if any part of the script is based on GAI-produced material, so that the writer's compensation, credit and separated rights will not be affected by the use of GAIproduced material." Before the writers went on strike, the studios rejected the proposal and countered by "offering annual meetings to discuss advancements in technology," according to the WGA.Read more of this story at Slashdot.

Huawei is accused by a semiconductor manufacturers association of setting up secret chip-making facilities in China to evade U.S. sanctions. The Guardian reports: The Chinese tech firm moved into chip production last year and was receiving an estimated $30 billion in state funding from the government, the Washington-based Semiconductor Industry Association was quoted as saying by Bloomberg, adding that Huawei had acquired at least two existing plants and was building three others. If Huawei is constructing facilities under names of other companies, as the Semiconductor Industry Association alleges, then it may be able to circumvent U.S. government restrictions to indirectly purchase American chip-making equipment, according to Bloomberg.Read more of this story at Slashdot.

An anonymous reader quotes a report from Ars Technica: A newly discovered zeroday in the widely used WinRAR file-compression program has been under exploit for four months by unknown attackers who are using it to install malware when targets open booby-trapped JPGs and other innocuous inside file archives. The vulnerability, residing in the way WinRAR processes the ZIP file format, has been under active exploit since April in securities trading forums, researchers from security firm Group IB reported Wednesday. The attackers have been using the vulnerability to remotely execute code that installs malware from families including DarkMe, GuLoader, and Remcos RAT. From there, the criminals withdraw money from broker accounts. The total amount of financial losses and total number of victims infected is unknown, although Group-IB said it has tracked at least 130 individuals known to have been compromised. WinRAR developers fixed the vulnerability, tracked as CVE-2023-38831, earlier this month. "By exploiting a vulnerability within this program, threat actors were able to craft ZIP archives that serve as carriers for various malware families," Group-IB Malware Analyst Andrey Polovinkin wrote. "Weaponized ZIP archives were distributed on trading forums. Once extracted and executed, the malware allows threat actors to withdraw money from broker accounts. This vulnerability has been exploited since April 2023." It's recommended that you update to version 6.23 before using WinRAR again.Read more of this story at Slashdot.

The G20 poured record levels of public money into fossil fuels last year despite having promised to reduce some of it, a report has found. The Guardian: The amount of public money flowing into coal, oil and gas in 20 of the world's biggest economies reached a record $1.4tn in 2022, according to the International Institute for Sustainable Development (IISD) thinktank, even though world leaders agreed to phase out aoeinefficienta fossil fuel subsidies at the Cop26 climate summit in Glasgow two years ago. The report comes ahead of a meeting of G20 countries in Delhi next month that could set the tone for the next big climate conference, which takes place in the United Arab Emirates in November. It is crucial that leaders put fossil fuel subsidies on the agenda, said Tara Laan, a senior associate with the IISD and lead author of the study. "These figures are a stark reminder of the massive amounts of public money G20 governments continue to pour into fossil fuels -- despite the increasingly devastating impacts of climate change." Fossil fuels release pollutants when burned that heat the planet and make extreme weather more violent. They also dirty the air with toxins that damage people's lungs and other organs. Scientists estimate the air pollution from fossil fuels kills between 1 and 10 million people each year. But beyond the overlooked costs to society, governments have lowered prices further by supporting fossil fuel producers and their customers with public money. The report found G20 governments last year provided fossil fuels $1tn in subsidies, $322bn in investments by state-owned enterprises and $50bn in loans from public finance institutions.Read more of this story at Slashdot.

The hazards are many. And they seem to come in all forms. From a report: The southwestern U.S. is reeling from record rainfall and extensive flooding from a rare tropical storm. Much of the central and southern parts of the country are in the grips of yet another oppressive heat wave. Nearly two weeks after catastrophic wildfires devastated the Hawaiian island of Maui, more fires are raging in the Pacific Northwest. And after a quiet start to this year's Atlantic hurricane season, activity in the basin is ramping up. All told, the various extremes are making for a turbulent week in nearly every corner of the country. Climate scientists also say it's an all-too-real look at how global warming increases the risks -- and consequences -- of the deadly events. "We're looking at a multi-hazard situation, where we're being hit by a string of different events over a short period of time," said Gonzalo Pita, an associate scientist and expert in disaster risk modeling at Johns Hopkins University. "It's like a double or triple whammy, and when they happen frequently or at the same time, the negative effects are compounded." While it's sometimes difficult to measure the exact role of climate change in any particular weather event, scientists know that global warming is having an overall effect on the frequency and severity of such events. Studies have shown, for instance, that heat waves and drought are more likely in a warming world. Dry conditions subsequently increase the risk of wildfires. Similarly, warmer-than-usual oceans are a key ingredient for tropical storms and hurricanes to form. A warmer atmosphere can also hold more moisture, making the storms rainier and likelier to cause flooding. Those types of compounding risks will be on full display this week. Tropical Storm Hilary on Sunday became the first to hit Southern California in 84 years, dumping record rain over the region and causing widespread flash flooding. Though Hilary has weakened into a post-tropical cyclone, 26 million people were still under flood alerts Monday across parts of California, Arizona, Utah, Nevada, Oregon and Idaho.Read more of this story at Slashdot.

An anonymous reader shares a report: When Jen Easterly, the director of the Cybersecurity and Infrastructure Security Agency, addresses the annual Chemical Security Summit in Arlington later this month, she'll be without a Big Stick she once wielded over the industry. Safeguarding the cybersecurity and physical security of 3,242 high-risk chemicals facilities across the country is one of CISA's critical responsibilities. Congress has renewed this authority, dubbed the Chemical Facility Anti-Terrorism Standards (CFATS), several times since enacting it in 2007. But on July 28, lawmakers for the first time allowed it to expire -- due in large part to the objections of a single senator. And there's no indication of when they might renew it after the Senate returns from recess in September. As a result, the risk that terrorists could weaponize dangerous chemicals produced in some of these facilities has increased, according to a senior chemical security official with CISA, who requested anonymity in order to share sensitive details about the effects of the lapse. Some of the 322 most sensitive chemicals can be used to make bombs or be released as toxic clouds, according to the official, who added that a direct attack on a facility could cause an explosion comparable to a nuclear blast. That's not all. Without the CFATS rule, CISA also effectively has no idea who is visiting facilities or if they are stockpiling dangerous chemicals, according to the official. Until Congress renews the rule, the agency can't send inspectors to the 160 facilities they typically visit every month. The agency can also no longer enforce penalties on facilities that violate its safety standards. At least one high-risk facility that was paying the agency's $40,000-a-day fine for failing to redress concerns (after receiving a warning) has stopped paying, according to the official.Read more of this story at Slashdot.

Two founders of Tornado Cash, the widely known Russian cryptocurrency mixer, have been charged with laundering more than $1 billion in criminal proceeds. From a report: In a newly unsealed indictment, Roman Storm and Roman Semenov have both been accused of sanctions violations and laundering money through Tornado Cash, including hundreds of millions of dollars for the Lazarus Group, a sanctioned North Korean state-backed hacking group. Charges in the indictment include conspiring to commit money laundering, conspiracy to commit sanctions violations and conspiracy to operate an unlicensed money transmitting business. Storm was arrested Wednesday in Washington state, according to a statement from the Justice Department, but Semenov, a Russian national, remains at large. The third co-founder, Alexey Pertsev, who is not mentioned in this action, faces trial in Amsterdam over his involvement with Tornado Cash. "Roman Storm and Roman Semenov allegedly operated Tornado Cash and knowingly facilitated this money laundering," said U.S. Attorney Damian Williams, adding, "While publicly claiming to offer a technically sophisticated privacy service, Storm and Semenov in fact knew that they were helping hackers and fraudsters conceal the fruits of their crimes." Further reading: Tornado Cash Co-founder Reports Being Kicked Off GitHub as Industry Reacts To Sanctions (2022);Coinbase Employees and Ethereum Backers Sue US Treasury Over Tornado Cash Sanctions (2022).Read more of this story at Slashdot.

Two UK teenagers accused of being key members of the notorious hacking group Lapsus$, behind attacks on companies including Nvidia, Rockstar Games, and Uber, were convicted of their crimes by a London jury Wednesday. From a report: Arion Kurtaj, 18, and a 17-year-old male, who can't be identified, were found to have carried out a number of offenses including serious computer misuse, blackmail and fraud against BT Group's EE network and Nvidia. Kurtaj was also separately accused of hacks into Uber, Rockstar's Grand Theft Auto game, and fintech firm Revolut. The Southwark Crown Court jury only needed to come to a decision on whether Kurtaj was liable for the crimes after he was found by the judge to be unfit to stand trial because of a complex medical condition. The jury found him liable for all 12 charges. The 17-year-old was found guilty of hacking, fraud and blackmail against Nvidia and cleared over two other counts against EE. He had previously plead guilty to two charges relating to the BT hacks. Lapsus$ are an international bunch of loosely connected online extortionists.Read more of this story at Slashdot.

An anonymous reader shares a report: Zoom CEO Eric Yuan told employees this month that the company was making the surprising decision to send some workers back to the office regularly because its flagship remote-work product didn't allow employees to build as much trust or be as innovative as in the office, according to a leaked meeting recording viewed by Insider. Zoom, one of the main enablers and beneficiaries of remote work, told employees living within 50 miles of a Zoom office that they must work there at least two days a week. The top reason for the mandate, Yuan said at the August 3 meeting, is that it's difficult for employees to get to know each other and build trust remotely. "In our early days, we all knew each other," Yuan said. "Over the past several years, we've hired so many new 'Zoomies' that it's really hard to build trust." He added: "Trust is a foundation for everything. Without trust, we will be slow." The second reason, he said, is that Zoom doesn't enable employees to have the conversations and debates that lead to innovation. "Quite often, you come up with great ideas, but when we are all on Zoom, it's really hard," Yuan said. "We cannot have a great conversation. We cannot debate each other well because everyone tends to be very friendly when you join a Zoom call."Read more of this story at Slashdot.

Cryptocurrency assets have amplified rather than reduced financial risks in less developed economies, and regulators will need to treat them in the same way they oversee other assets, some of the world's most powerful central banks have warned. From a report: Novel solutions to payments challenges should not be classified as 'dangerous' simply because they are different, the Bank for International Settlements said on Tuesday. However the global central banking body added that the appeal of crypto was "illusory," in a paper published on approaches to regulation. The Consultative Group of Directors of Financial Stability, which includes representatives from central banks of the US, Argentina, Brazil, Canada, Chile and Mexico, said crypto had been promoted as a low-cost payment solution and substitute for national currencies in countries with high inflation or high exchange rate volatility. "However, crypto assets have so far not reduced but rather amplified the financial risks in less developed economies. Therefore, they should be assessed from a risk and regulatory perspective like all other assets," it said in a 50-page report. Watchdogs including the IMF and the Bank for International Settlements have been charting the evolving financial stability risks from the cryptocurrency market as it ballooned from a nascent industry to one whose value peaked at $2.9tn in November 2021.Read more of this story at Slashdot.

Sony is officially launching its portable PlayStation later this year, the PlayStation Portal remote player. The handheld device will stream PS5 games over Wi-Fi and features an eight-inch LCD screen running at 1080p resolution at 60fps. Sony says the PlayStation Portal will be available later this year priced at $199.99. From a report: "PlayStation Portal will connect remotely to your PS5 over Wi-Fi, so you'll be able to swiftly jump from playing on your PS5 to your PlayStation Portal," says Hideaki Nishino, senior vice president of platform experience at Sony Interactive Entertainment. "PlayStation Portal can play supported games that are installed on your PS5 console and use the Dualsense controller." The PlayStation Portal features prominent controllers on each side that look very much like Sony's PS5 DualSense controllers. They support adaptive triggers and haptic feedback, so PS5 games will feel similar to using a dedicated DualSense controller. The PlayStation Portal will also be capable of playing media, as the homescreen has a dedicated section for it as it's mirroring your PS5. You won't be able to run anything locally though, so if you don't have Wi-Fi then you're out of luck.Read more of this story at Slashdot.

India has become the first country to land a spacecraft on the moon's south pole. It launched Chandrayaan-3 in mid-July, with the spacecraft entering the moon's orbit on Aug. 5. Earlier this week, Russia crashed its Luna-25 rocket in the same lunar region. From a report: It's notoriously difficult to land a rover on the moon. Russia's Luna-25 crashed while making an attempt just this week, while Japanese company ispace failed to land an unmanned lander in April.a 1 Since the moon has no atmosphere, landers can't just softly touch down on the lunar surface. And, without GPS capabilities, scientists rely on the lander's computers to accurately identify where the spacecraft will touch down. India is only the fourth country to pull off a moon landing, behind the U.S., China, and Russia. The nation's lunar aspirations are part of a push by Prime Minister Narendra Modi to garner investments for private space exploration and satellite launches. Speaking at the rocket's launch in July, Modi heralded a "new chapter" in India's space program.Read more of this story at Slashdot.

An anonymous reader quotes a report from the Associated Press: Pennsylvania's Supreme Court ruled Tuesday that the state police can't hide from the public its policy on how it monitors social media. Advocates for civil liberties cheered the decision. The law enforcement agency had argued that fully disclosing its policy for using software to monitor online postings may compromise public safety. All four Democratic justices supported the majority decision, which said the lower Commonwealth Court went beyond its authority in trying to give the state police another attempt to justify keeping details of the policy a secret. Tuesday's order appears to end a six-year legal battle. Justifying what the majority opinion described as heavy or complete redactions on every page of the nine-page regulation, the head of the state police's bureau of criminal investigations argued that greater transparency about the policy would make its investigations less effective. The state Office of Open Records held a private review of the blacked out material and and ruled that making the policy public would not be likely to harm investigations, calling the social media policy processes strictly internal and administrative in nature. Redacted sections addressed the use of open sources, what approval is required, when to go undercover and use an online alias and how to verify information. State police also blacked out the entire section on using social media for employment background investigations. A panel of three Republican Commonwealth Court judges reversed the Office of Open Records' ruling that the policy should be disclosed without redactions, saying in May 2018 that the state police investigations chief based his analysis about the risk of exposure on his own extensive experience. The majority decision issued Tuesday said Commonwealth Court should not have given the state police a new opportunity to lay out the supposed public safety risks. The majority ruled that Pennsylvania's Right-to-Know Law does not permit Commonwealth Court to order additional fact-finding not sought by state police. Andrew Christy, a lawyer with the ACLU of Pennsylvania, said the ruling "sort of puts law enforcement on the same playing field as all government agencies. If they have a legal justification to keep something secret, then they have to put forth sufficient evidence to justify that." "Ultimately that relies on the voters understanding what law enforcement is doing so that then, through their elected representatives, they can rein them in when they're acting in a way that doesn't comport with what the public wants," Christy said.Read more of this story at Slashdot.

New York City, home to more than a million students in its school system, is the biggest school district in the U.S. -- and now allows any student to enroll virtually in the aftermath of the COVID-19 pandemic. Time reports: Dubbed Virtual Innovators Academy, there are 17 teachers for about 200 students enrolled in the 2023-2024 school year for sophomore and freshman years. Each year, another grade level will be added, and the school's funding comes from the city and state, just like other public schools. Students meet in-person for required state exams and for monthly social gatherings like arcade games at Dave & Busters or seeing a Broadway show. But many of the most popular extracurriculars are done from home, says Virtual Innovators Academy principal Terri Grey, like esports and flying drones. [...] And it isn't just in New York: school districts in Utah, Georgia, California, and elsewhere have also launched permanent virtual schools. Concerns remain about the effectiveness of virtual school. Critics worry about the lack of in-person social interaction during crucial developmental years, and about whether teachers can educate as effectively through a screen. But administrators behind the nation's burgeoning virtual schools say they have studied what works and what doesn't from remote-schooling during the pandemic when setting up these communities. Every morning, students at Virtual Innovators Academy meet in small groups with a teacher advisor to talk about how they're doing and give them time to wake up in the morning and connect with other classmates. There's less emphasis on multiple choice tests, which proved harder to administer online, and more emphasis on research projects. "Too many people judge virtual instruction as if it were the emergency roadside online instruction that happened as a result of the pandemic," says Anthony Godfrey, who helps oversee the K-12 Jordan Virtual Learning Academy in Utah. "This is something very different. This is a carefully thought out, very intentional way of providing a unique and effective means of instruction." [...] But for all the proponents of virtual schooling, there are critics who worry about what's being lost behind the computer screen. [...] Unstructured, spontaneous conversations are often the most memorable parts of school, he argues; students might work side-by-side, help each other with homework, and also socialize in between classes. In virtual school, "How do you create space for bumping into somebody in the hall?" [wonders Nathan Holbert, a researcher at Teachers College, Columbia University, who studies virtual learning]. "I don't know that you can."Read more of this story at Slashdot.

A two-year human trial conducted by James Cook University (JCU) has concluded, demonstrating positive results using low-dose human hookworm therapy to treat chronic conditions, particularly in relation to type 2 diabetes. New Atlas reports: [O]f the 24 participants who received worms, when offered a dewormer at the end of the second year of the trial, with the option to stay in the study for another 12 months, only one person chose to kill off their gut buddies -- and it was only because they had an impending planned medical procedure. "All trial participants had risk factors for developing cardiovascular disease and type 2 diabetes," said Dr Doris Pierce, from JCU's Australian Institute of Tropical Health and Medicine (AITHM). "The trial delivered some considerable metabolic benefits to the hookworm-treated recipients, particularly those infected with 20 larvae." In this double-blinded trial, 40 participants aged 27 to 50, with early signs of metabolic diseases, took part. They received either 20 or 40 microscopic larvae of the human hookworm species Necator americanus; another group took a placebo. As an intestinal parasite, the best survival skill is to keep the host healthy, which will provide a long-term stable home with nutrients 'on tap.' In return, these hookworms pay the rent in the form of creating an environment that suppresses inflammation and other adverse conditions that can upset that stable home. While the small, round worms can live for a decade, they don't multiply unless outside the body, and good hygiene means transmission risk is very low. As for the results, those with 20 hookworms saw a Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) level drop from 3.0 units to 1.8 units within the first year, which restored their insulin resistance to a healthy range. The cohort with 40 hookworms still experienced a drop, from 2.4 to 2.0. Those who received the placebo saw their HOMA-IR levels increase from 2.2 to 2.9 during the same time frame. "These lowered HOMA-IR values indicated that people were experiencing considerable improvements in insulin sensitivity -- results that were both clinically and statistically significant," said Dr Pierce. Those with worms also had higher levels of cytokines, which play a vital role in triggering immune responses. The study was published in the journal Nature Communications.Read more of this story at Slashdot.

An anonymous reader quotes a report from The Guardian: Methane is a potent greenhouse gas emitted from energy (natural gas and petroleum systems), industry, agriculture, land use and waste management activities. Now a group of researchers from California University Long Beach are proposing a method of removing methane by using a group of bacteria known as methanotrophs to naturally convert methane to carbon dioxide and biomass. All the bacteria in this group "'eat' methane, removing it from air and converting part of it to cells as a source of sustainable protein," according to the lead researcher, Mary E Lidstrom. Lidstrom's team have found a strain of bacteria within this group called methylotuvimicrobium buryatense 5GB1C that can remove methane efficiently even when it is present in lower amounts. If it became widespread, the technology has the potential to help slow global heating, the researchers said. Typically, this group of bacteria thrive in environments with high levels of methane (between 5,000 and 10,000 parts per million (ppm)). The normal concentrations in our atmosphere have much lower levels of only about 1.9 ppm of methane. But certain areas such as landfills, rice fields and oilwells emit higher concentrations of about 500 ppm. "Bacteria that rapidly eat methane at the higher concentrations found around cattle herds, etc could make a huge contribution to cutting methane emissions, especially from tropical agriculture," said Euan Nisbet, professor of Earth sciences at Royal Holloway, University of London, commenting on the findings of the study. The strain's high methane consumption rate is probably due to a low energy requirement and greater attraction for methane - more than five times more than that of other bacteria, according to the study. "The bacteria oxidise the methane to CO2 (a much less powerful greenhouse gas) and so you can even use the exhaust to pump into greenhouses and grow tomatoes," said Nisbet. "The biggest barrier to implementation now is technical: we need to increase the methane treatment unit 20-fold. If we can achieve that, then the biggest barriers become investment capital and public acceptance. We believe we could have field pilots tested within three to four years, and scale up would then depend on investment capital and commercialization," said Lidstrom. The study has been published in the journal Proceedings of the National Academy of Sciences (PNAS).Read more of this story at Slashdot.

Switzerland's MeteoSuisse reported another measure of record summer heat Monday when its weather balloon climbed to a record-high 17,400 feet before reaching what it calls the zero-degree line. UPI reports: The zero-degree line, which is the altitude at which the temperature falls below freezing, is considered a key meteorological marker, particularly in mountainous regions, has been climbing and set a record in 2022. "The Payerne, [Switzerland] radiosounding this night from August 20 to 21, 2023 measured the 0C isothermal 5,298 meters, which is a record since the start of measurements in 1954," MeteoSuisse said in a translated social media post. The weather service said the zero-degree line "affects vegetation, the snow line and the water cycle so has a considerable impact on the habitats of humans, animals and plants alike." The zero-degree line averaged 8,432 feet from 1991 to 2020, with a high of about 13,123 feet in the summer. "Anthropogenic climate change has caused the altitude of the zero-degree line to rise significantly in every season," MeteoSuisse said.Read more of this story at Slashdot.

According to Windows Central, Microsoft might be bringing AI capabilities to a handful of Windows 11 apps, including Photos, Snipping Tool, and Paint. "Some of this functionality will require dedicated hardware, such as an NPU (neural processing unit) or VPU (vision processing unit,) while others may not," notes the report. From the report: For the Photos app, Microsoft is working on an AI functionality that would allow the app to identify objects or people in photos and enable the ability to cut out and paste those elements elsewhere. This is a functionality that iOS and Android have had for some time, so it's no surprise to hear that Microsoft is also working to bring it to Windows. Regarding the Snipping Tool, my sources say the company wants to incorporate OCR (optical character recognition) technology to enable Windows to identify text in screenshots for faster clipboard copying. Microsoft is also working on bringing OCR to the Camera app, allowing users to select text in a photo taken on the device. Lastly, my sources say Microsoft has also been experimenting with bringing generative AI to the Windows 11 Paint app. Users could ask Paint to create a canvas based on criteria set out by the user, similar to how Bing Image Creator currently works. Sources say the Paint AI integration will be based on that same Bing technology.Read more of this story at Slashdot.

Long-time Slashdot reader UnknowingFool writes: In 2021, Craig Wright sued 12 bitcoin developers who refused help him recover 111,000 bitcoins he claimed were lost in a hack. His company, Tulip Trading, wanted the developers to put in a backdoor mechanism in bitcoin that would override the ownership of the coins, arguing it was the developers "fiduciary duty" to assist him. The developers allege (PDF) that Tulip and Wright never owned the coins and the evidence of ownership provided is "fabricated." Tulip Trading "never owned the digital assets and has commenced this claim fraudulently and in reliance on fabricated documents," the developers' lawyers said in a statement. "Dr. Wright has a long history of fraud, forgery, and dishonesty ... [and is using] the English courts as an instrument of fraud."Read more of this story at Slashdot.

An anonymous reader quotes a report from CBC.ca: Prime Minister Justin Trudeau blasted social media giant Meta on Monday over its decision to block local news as wildfires continue to force thousands of Canadians from their homes. "Right now in an emergency situation, where up-to-date local information is more important than ever, Facebook is putting corporate profits ahead of people's safety, ahead of quality local journalism. This is not the time for that," he said during a stop at the Island Montessori Academy in Cornwall, P.E.I. on Monday morning. "It is so inconceivable that a company like Facebook is choosing to put corporate profits ahead of ensuring that local news organizations can get up-to-date information to Canadians and reach them where Canadians spend a lot of their time -- online, on social media, on Facebook." Meta, the parent company of Facebook and Instagram, has blocked Canadians from viewing news from Canadian outlets in response to the Liberal government passing its Online News Act, Bill C-18, in June. Google has threatened similar action. The law forces large social media platforms to negotiate compensation for Canadian news publishers when their content is shared. As a result, content from news providers in the North -- including CBC, the local newspaper The Yellowknifer and digital broadcaster Cabin Radio -- is being blocked and people can't access or share information from news sources on Facebook and Instagram, two of the most popular social media sites. In a statement sent to CBC News last week, the company said it's sticking to its position. It also said government sites and other sources that disseminate information aren't subject to the ban. "This is Facebook's choice," said Trudeau. "We're simply saying that in a democracy, quality local journalism matters. And it matters now more than ever before, when people are worried about their homes, worried about communities, worried about the worst summer for extreme weather events we've had in a long, long time." Meanwhile, Meta spokesperson David Troya-Alvarez said: "People in Canada are able to use Facebook and Instagram to connect to their communities and access reputable information, including content from official government agencies, emergency services and non-governmental organizations." Meta says it has activated a "Safety Check" feature that allows users to mark on their profile they're safe from the wildfires.Read more of this story at Slashdot.

IBM is introducing the watsonx Code Assistant for Z, a tool that uses generative AI to translate COBOL code to Java. This tool is set to be available in Q4 2023 and aims to speed up the translation of COBOL to Java on IBM's Z mainframes. The Register reports: According to IBM, there are billions of lines of COBOL code out there as potential candidates for modernization (a report last year estimated the total figure at 775-850 billion lines). For this reason, the generative AI features in watsonx Code Assistant for Z are intended to help developers to assess and determine the code most in need of modernization, allowing them to more speedily update large applications and focus on critical tasks. IBM wants to provide tooling for each step of the modernization process, starting with its Application Discovery and Delivery Intelligence (ADDI) inventory and analysis tool. Other steps include refactoring business services in COBOL, transforming the code to Java code, and then validating the resulting outcome with the aid of automated testing. The resulting Java code emitted by watsonx Code Assistant for Z will be object-oriented, but will still interoperate with the rest of the COBOL application IBM claimed, as well as with key services such as CICS, IMS, DB2, and other z/OS runtimes.Read more of this story at Slashdot.

A draft agreement between TikTok and the Committee on Foreign Investment in the United States (CFIUS) to avoid a ban would have given U.S. agencies unprecedented access to TikTok's facilities and servers. "Many of the concessions the government asked of TikTok look eerily similar to the surveillance tactics critics have accused Chinese officials of abusing," reports Gizmodo. "To allay fears the short-form video app could be used as a Chinese surveillance tool, the federal government nearly transformed it into an American one instead." The draft of the deal was obtained by Forbes. From a report: Forbes reports that the draft agreement, dated Summer 2022, would have given the US government agencies like the Department of Justice and Department of Defense far more access to TikTok's operations than that of any other social media company. The agreement would let agencies examine TikTok's US facilities, records, and servers with minimal prior notice and veto the hiring of any executive involved with leading TikTok US data security organization. It would also let US agencies block changes to the app's terms of service in the US and order the company to subject itself to various audits, all on TikTok's dime, per Forbes. In extreme cases, the agreement would allow government organizations to demand TikTok temporarily shut off functioning in the U.S.. The draft document, which Gizmodo could not independently verify, is reportedly around 100 pages long and contains comments sent between attorneys representing ByteDance, TikTok's Chinese-owned parent company, and CFIUS. The agreements, if accepted as written at the time, would open TikTok's U.S. operations up to supervision by a number of external third-party auditors and source code inspectors. ByteDance leaders, whom US lawmakers and whistleblowers have accused of maintaining close connections with The Chinese Communist Party, would be excluded from some security-related decisions involving the US version of the app. Provisions described in the guidelines weren't always agreed on by both parties. In several instances, according to Forbes, TikTok's attorneys pushed back against terms that would let the government alter what types of user data ByteDance employees could view. Another point of disagreement emerged when the government reportedly asked for limitless veto power over TikTok's future contracts. At one point, TikTok reportedly altered language that would have allowed government officials to demand changes to the apps recommendations algorithm if it promoted content the agencies disagreed with. A TikTok spokesperson said in a statement: "As has been widely reported, we've been working with CFIUS for well over a year to implement a national security agreement and have invested significant resources in implementing a firewall to isolate U.S. user data. Today, all new protected U.S. user data is stored in the Oracle Cloud Infrastructure in the U.S. with tightly controlled and monitored gateways. We are doing more than any peer company to safeguard U.S. national security interests."Read more of this story at Slashdot.

An anonymous reader quotes a report from Ars Technica: On Tuesday, Meta announced SeamlessM4T, a multimodal AI model for speech and text translations. As a neural network that can process both text and audio, it can perform text-to-speech, speech-to-text, speech-to-speech, and text-to-text translations for "up to 100 languages," according to Meta. Its goal is to help people who speak different languages communicate with each other more effectively. Continuing Meta's relatively open approach to AI, Meta is releasing SeamlessM4T under a research license (CC BY-NC 4.0) that allows developers to build on the work. They're also releasing SeamlessAlign, which Meta calls "the biggest open multimodal translation dataset to date, totaling 270,000 hours of mined speech and text alignments." That will likely kick-start the training of future translation AI models from other researchers. Among the features of SeamlessM4T touted on Meta's promotional blog, the company says that the model can perform speech recognition (you give it audio of speech, and it converts it to text), speech-to-text translation (it translates spoken audio to a different language in text), speech-to-speech translation (you feed it speech audio, and it outputs translated speech audio), text-to-text translation (similar to how Google Translate functions), and text-to-speech translation (feed it text and it will translate and speak it out in another language). Each of the text translation functions supports nearly 100 languages, and the speech output functions support about 36 output languages.Read more of this story at Slashdot.

A piece of space debris being monitored by the European Space Agency as part of a mission to remove trash from space was hit by another piece of debris, splintering the object into more pieces. From a report: ESA confirmed Tuesday that the US's 18th Space Defense Squadron, which tracks objects in orbit, spotted a number of new pieces in the vicinity of a payload adapter named VESPA that the agency had planned to pluck from space. The most likely cause of those new fragments is "the hypervelocity impact of a small, untracked object" ramming into VESPA, according to ESA. VESPA was left over from the launch of a European Vega rocket that took off from South America in 2013. It was part of a cone-shaped attachment used to deploy the rocket's satellite into orbit, and has been in Earth's orbit ever since. ESA said its new fragments don't pose much of a risk to any other spacecraft at the moment.Read more of this story at Slashdot.

Japan is to begin releasing wastewater from the wrecked Fukushima Daiichi nuclear power plant from Thursday, in defiance of opposition from fishing communities, China and some scientists. From a report: The prime minister, Fumio Kishida, said on Tuesday he had asked the plant's operator, Tokyo Electric Power (Tepco), "to swiftly prepare for the water discharge" in accordance with plans approved by nuclear regulators, adding that the release would begin on Thursday, "weather and ocean conditions permitting." Kishida has said that disposing of more than 1m tonnes of water being stored at the site was an essential part of the long and complex process to decommission the plant. The plan has caused controversy because the water contains tritium, a radioactive substance that can't be removed by the facility's water filtration technology. Hong Kong, an important market for Japanese seafood exports, has threatened restrictions. Leader John Lee said on Tuesday he strongly opposed the water plan, adding that he had instructed the city's government to "immediately activate" import controls on Japanese seafood. South Korea and China banned seafood imports from some areas of Japan after Fukushima Daiichi suffered a triple meltdown in the March 2011 triple disaster along the country's north-east coast. China remains strongly opposed, accusing Japan of treating the ocean like a "sewer." The South Korean government recently dropped its objections to the discharge, but opposition parties and many South Koreans are concerned about the impact the discharge will have on food safety.Read more of this story at Slashdot.

Once tooth decay has set in, all a dentist can do is fill the gap with an artificial plug -- a filling. But in a paper published in Cell, Hannele Ruohola-Baker, a stem-cell biologist at the University of Washington, and her colleagues offer a possible alternative. Economist: Stem cells are those that have the capacity to turn themselves into any other type of cell in the body. It may soon be possible, the researchers argue, to use those protean cells to regrow a tooth's enamel naturally. The first step was to work out exactly how enamel is produced. That is tricky, because enamel-making cells, known as ameloblasts, disappear soon after a person's adult teeth have finished growing. To get round that problem, the researchers turned to samples of tissue from human foetuses that had been aborted, either medically or naturally. Such tissues contain plenty of functioning ameloblasts. The researchers then checked to see which genes were especially active in the enamel-producing cells. Tooth enamel is made mostly of calcium phosphate, and genes that code for proteins designed to bind to calcium were particularly busy. They also assessed another type of cell called odontoblasts. These express genes that produce dentine, another type of hard tissue that lies beneath the outer enamel. Armed with that information, Dr Ruohola-Baker and her colleagues next checked to see whether the stem cells could be persuaded to transform into ameloblasts. The team devised a cocktail of drugs designed to activate the genes that they knew were expressed in functioning ameloblasts. That did the trick, with the engineered ameloblasts turning out the same proteins as the natural sort. A different cocktail pushed the stem cells to become odontoblasts instead. Culturing the cells together produced what researchers call an organoid -- a glob of tissue in a petri dish which mimics a biological organ. The organoids happily churned out the chemical components of enamel. Having both cell types seemed to be crucial: when odontoblasts were present alongside ameloblasts, genes coding for enamel proteins were more strongly expressed than with ameloblasts alone. For now, the work is more a proof of concept than a prototype of an imminent medical treatment. The next step, says Dr Ruohola-Baker, is to try to boost enamel production even further, with a view to eventually beginning clinical trials. The hope is that, one day, medical versions of the team's organoids could be used as biological implants, to regenerate a patient's decayed teeth.Read more of this story at Slashdot.

VMware on Tuesday said it has developed a new set of software tools in partnership with Nvidia aimed at businesses which want to develop generative artificial intelligence in their own data centers rather than the cloud. From a report: VMware, which is close to being acquired by chip firm Broadcom in a $69 billion deal, makes software that corporations use to run their privately owned data centers. For more than two decades, VMware's tools have been used by businesses to divvy up the computing power in central processor chips, which are the brains of traditional servers. On Tuesday, the company released a new set of tools help designed to manage Nvidia chips, which dominate the market for AI systems that can read and write text in human-like ways.Read more of this story at Slashdot.

Amazon has launched a new Fire TV Channels app, giving Fire TV customers access to over 400 free ad-supported TV (FAST) channels, including ABC News, CBS Sports, Fox Sports, MLB, Martha Stewart and more. From a report: Alongside the app launch, the company also announced new content providers, such as Variety, Rolling Stone, The Hollywood Reporter, Billboard, GameSpot, Looper and Funny or Die, among others. Amazon will continue adding more FAST channels over time, the company wrote in the press release. First introduced in May, Fire TV Channels are available on Fire TV-branded smart TVs and streaming devices. The new app offers a single destination for users to quickly access thousands of live and on-demand titles across sports, news, entertainment, cooking, gaming and more. Customers can find the Fire TV Channels app via the "Your Apps and Channels" section, the Free icon in the navigation bar or labeled content rows throughout the Fire TV user interface. They also can ask Alexa to "Play Fire TV Channels." No downloads are required to watch FAST channels on Fire TV.Read more of this story at Slashdot.

A decade after data analytics promised to revolutionize agriculture, most farmers still aren't using data tools or specialized software, and of those who do, many are swamped in a deluge of data. From a report: In 2013, seed and pesticide giant Monsanto acquired agriculture-data firm Climate Corporation for $1 billion, helping spur the industry's mania for data-driven farming. The hope was that by outfitting farmers with software and tools capable of ingesting and analyzing troves of data on things from weather patterns to soil conditions, they could more efficiently use their land. Many are still waiting for the technology to pay off. In the U.S., less than half of farmers surveyed by consulting firm McKinsey are using farm management software, and 25% are using remote-sensing and precision agriculture hardware. That software is a foundational technology in enabling the autonomous machinery and AI-enabled equipment of the future, analysts say, and unless farmers start using it, some will be left behind in the next decade of farm innovation. At the moment, 3% of American farmers said they plan to adopt software or precision agriculture hardware over the next two years, according to McKinsey. Certain tools can automatically gather data from internet-connected farm equipment, but others require farmers to manually enter the information. For a specific field, for instance, that could total over a dozen crop-protection products and multiple seeds. Even those who are using the tech say they can find it difficult to draw useful conclusions from it. "We're collecting so much data that you're almost paralyzed with having to analyze it all," said David Emmert, a corn and soybean farmer in West Central Indiana who works about 4,300 acres. [...] The first generation of digital farming tools also wasn't easy for farmers to use. Software was slow, interfaces were complex and difficult to manage. "The industry does need to step up a little bit on continuing to improve the customer experience," said David Fiocco, a McKinsey partner focused on agriculture. In recent years, big tech vendors like Microsoft, Amazon and Google have begun tailoring their cloud-computing, data and artificial-intelligence services to agriculture, bringing along expertise that could help address complications that have long plagued farm data management and analytics.Read more of this story at Slashdot.

The United Nations' proposed Global Digital Compact will exclude technical experts as a distinct voice in internet governance, ignoring their enormous contributions to growing and sustaining the internet, according to ICANN and two of the world's regional internet registries. From a report: The Global Digital Compact is an effort to "outline shared principles for an open, free and secure digital future for all." The UN hopes the compact will address issues such as digital inclusion, internet fragmentation, giving individuals control over how their data is used, and making the internet trustworthy "by introducing accountability criteria for discrimination and misleading content." But ICANN, the Asia Pacific Network Information Centre (APNIC), and the American Registry for Internet Numbers (ARIN) worry that recent articulations of the Compact suggest it should use a tripartite model for digital cooperation with three stakeholder groups: the private sector, governments, and civil society. That's dangerous, ICANN and co argue, because technical stakeholders would lose their distinct voice. They've therefore co-signed and published a document criticizing the Compact as it stands today. "The technical community is not part of civil society and it has never been," the document states, citing outcomes of the World Summit of the Information Society (WSIS) -- a UN event staged in 2003 and 2005 that defined a multi-stakeholder internet governance framework. 2015's WSIS+10 event affirmed that strategy. "This model excludes the technical community as a distinct component, and overlooks the unique and essential roles played by that community's members separately and collectively," DNS overlord ICANN and the registries added.Read more of this story at Slashdot.

Microsoft is restructuring its proposed Activision Blizzard deal to transfer cloud gaming rights for current and new Activision Blizzard games to Ubisoft. From a report: The transfer of rights is designed to appease regulators in the UK that are concerned about the impact Microsoft's proposed $68.7 billion deal will have on cloud gaming competition. The restructured deal has triggered a new regulatory investigation in the UK that could last until October 18th. "To address the concerns about the impact of the proposed acquisition on cloud game streaming raised by the UK Competition and Markets Authority, we are restructuring the transaction to acquire a narrower set of rights," says Microsoft president Brad Smith. "This includes executing an agreementaeffective at the closing of our merger that transfersathe cloud streaming rights for all current and new Activision Blizzard PC and console games released over the next 15 years to Ubisoft Entertainment SA, a leading global game publisher. The rights will be in perpetuity." This restructured deal means that if Microsoft does close its proposed acquisition, then it will not be able to release Activision Blizzard games exclusively on Xbox Cloud Gaming.Read more of this story at Slashdot.