An anonymous reader quotes a report from TechCrunch: Before 16-year-old Adam Raine died by suicide, he had spent months consulting ChatGPT about his plans to end his life. Now, his parents are filing the first known wrongful death lawsuit against OpenAI, The New York Times reports. Many consumer-facing AI chatbots are programmed to activate safety features if a user expresses intent to harm themselves or others. But research has shown that these safeguards are far from foolproof. In Raine's case, while using a paid version of ChatGPT-4o, the AI often encouraged him to seek professional help or contact a help line. However, he was able to bypass these guardrails by telling ChatGPT that he was asking about methods of suicide for a fictional story he was writing. OpenAI has addressed these shortcomings on its blog. "As the world adapts to this new technology, we feel a deep responsibility to help those who need it most," the post reads. "We are continuously improving how our models respond in sensitive interactions." Still, the company acknowledged the limitations of the existing safety training for large models. "Our safeguards work more reliably in common, short exchanges," the post continues. "We have learned over time that these safeguards can sometimes be less reliable in long interactions: as the back-and-forth grows, parts of the model's safety training may degrade."

Read more of this story at Slashdot.

Anthropic reached a settlement with authors in a high-stakes copyright class action that threatened the AI company with potentially billions of dollars in damages. From a report: In a Tuesday filing in the US Court of Appeals for the Ninth Circuit, both sides asked the court to pause all proceedings while they finalize the deal. The parties signed a binding term sheet on Aug. 25 outlining the core terms of a proposed class settlement to resolve litigation brought by authors. "This historic settlement will benefit all class members," said the authors' counsel, Justin Nelson of Susman Godfrey LLP. "We look forward to announcing details of the settlement in the coming weeks." The case is one of several copyright actions brought against AI developers in courts around the country. Judge William Alsup of the US District Court for the Northern District of California had allowed the class action to proceed for authors whose books were contained in two pirate databases Anthropic downloaded.

Read more of this story at Slashdot.

The Michigan Supreme Court has drawn a firm line around digital privacy, ruling that police cannot use overly broad warrants to comb through every corner of a person's phone. From a report: In People v. Carson, the court found [PDF] that warrants for digital devices must include specific limitations, allowing access only to information directly tied to the suspected crime. Michael Carson became the focus of a theft investigation involving money allegedly taken from a neighbor's safe. Authorities secured a warrant to search his phone, but the document placed no boundaries on what could be examined. It permitted access to all data on the device, including messages, photos, contacts, and documents, without any restriction based on time period or relevance. Investigators collected over a thousand pages of information, much of it unrelated to the accusation. The court ruled that this kind of expansive warrant violates the Fourth Amendment, which requires particularity in describing what police may search and seize.

Read more of this story at Slashdot.

Air pollution from oil and gas causes more than 90,000 premature deaths and sickens hundreds of thousands of people across the US each year, a new study shows, with disproportionately high impacts on communities of color. From a report: More than 10,000 annual pre-term births are attributable to fine particulate matter from oil and gas, the authors found, also linking 216,000 annual childhood-onset asthma cases to the sector's nitrogen dioxide emissions and 1,610 annual lifetime cancer cases to its hazardous air pollutants. The highest number of impacts are seen in California, Texas, New York, Pennsylvania and New Jersey, while the per-capita incidences are highest in New Jersey, Washington DC, New York, California and Maryland.

Read more of this story at Slashdot.

Swiss researchers have determined that fermentation degree controls beer foam stability after seven years of study published in Physics of Fluids. Triple-fermented Belgian beers maintained the longest-lasting foam while single-fermented lagers produced the shortest duration. The team tested six commercial beers including Westmalle Tripel, Tripel Karmeliet, and Swiss lagers Feldschlosschen and Chopfab. Surface viscosity dominated foam stability in single-fermented beers. Marangoni stresses from surface tension differences stabilized double- and triple-fermented beer foams. Lipid transfer protein 1 underwent progressive denaturation through successive fermentations. Single fermentation produced small round protein particles. Double fermentation created net-like protein structures. Triple fermentation broke proteins into hydrophobic and hydrophilic fragments that function as surfactants. ETH Zurich's Jan Vermant said breweries can now improve foam using these specific mechanisms rather than adjusting multiple factors simultaneously.

Read more of this story at Slashdot.

Google has integrated AI-powered language learning capabilities into its Translate app through a beta feature that generates customized lessons using its Gemini AI models. The Practice button allows English speakers to learn Spanish and French while Spanish, French, and Portuguese speakers can practice English. Users select their skill level and learning goals to receive tailored scenarios ranging from professional conversations to family interactions. The company also launched live translation for real-time conversations across 70 languages in the US, India, and Mexico. The feature creates AI-generated transcriptions and audio translations but does not replicate users' voices, the company told The Verge.

Read more of this story at Slashdot.

The Document Foundation, which operates the popular open source productivity suite LibreOffice, is positioning the suite's newest release, v25.8, as a strategic asset for digital sovereignty, targeting governments and enterprises seeking independence from foreign software vendors and cloud infrastructure. The Document Foundation released the update last week with zero telemetry architecture, full offline capability, and OpenPGP encryption for documents, directly addressing national security concerns about extraterritorial surveillance and software backdoors. The suite requires no internet access for any features and maintains complete transparency through open source code that governments can audit. Government bodies in Germany, Denmark, and France, alongside national ministries in Italy and Brazil, have deployed LibreOffice to meet GDPR compliance, national procurement laws, and IT localization mandates while eliminating unpredictable licensing costs from proprietary vendors. "It's time to own your documents, own your infrastructure, and own your future," the foundation wrote in a blog post.

Read more of this story at Slashdot.

French prosecutors have opened an investigation into the Australian video platform Kick over the death of a content creator during a live stream. From a report: Raphael Graven -- also known as Jean Pormanove -- was found dead in a residence near the city of Nice last week. He was known for videos in which he endured apparent violence and humiliation. The Paris prosecutor said the investigation would look into whether Kick knowingly broadcast "videos of deliberate attacks on personal integrity." The BBC has approached Kick for comment. A spokesperson for the platform previously said the company was "urgently reviewing" the circumstances around Mr Graven's death. The prosecutor's investigation will also seek to determine whether Kick complied with the European Union's Digital Services Act, and the obligation on platforms to notify the authorities if the life or safety of individuals is in question. In a separate announcement, France's minister for digital affairs, Clara Chappaz, said the government would sue the platform for "negligence" over its failure to block "dangerous content", according to the AFP news agency.

Read more of this story at Slashdot.

AT&T said Tuesday it would buy wireless licenses from EchoStar for $23 billion, after a years-long saga over what the latter would do with its vast spectrum holdings. From a report: EchoStar was reportedly under pressure from regulators and the White House to either start selling its spectrum or potentially lose it. The cash payment is almost three times the size of EchoStar's entire market capitalization. AT&T said the acquired spectrum covers "virtually every" U.S. market, and will let it speed up and expand the deployment of its home wireless Internet service, as well as continue the phase-out of traditional copper phone line service.

Read more of this story at Slashdot.

Entry-level workers in AI-exposed occupations have seen employment drop 13% since late 2022, according to Stanford University research analyzing millions of payroll records. The decline affects software developers, customer service representatives, and administrative assistants aged 22 to 25, while employment for older workers in the same roles continued growing. The study [PDF], based on ADP payroll data covering tens of thousands of firms, found the steepest drops in occupations where AI automates tasks rather than augments human capabilities. Among software developers aged 22-25, employment fell nearly 20% from its late 2022 peak. Workers in less AI-exposed fields like nursing saw employment growth across all age groups. The research controlled for firm-level effects and other economic factors, isolating AI's impact from broader trends like interest rate changes and pandemic-era hiring patterns.

Read more of this story at Slashdot.

An anonymous reader quotes a report from Reuters: AbbVie will buy an experimental depression drug from partner Gilgamesh Pharmaceuticals for up to $1.2 billion, the companies said on Monday, seeking to access a fast-growing market for psychedelic-based treatments. The deal is the latest in the more than $20 billion AbbVie has spent on acquisitions since 2023 for drugs that can drive growth as its flagship rheumatoid arthritis treatment, Humira, lost patent protection. The companies had signed a partnership last year to develop therapies for psychiatric disorders, with privately held Gilgamesh set to receive up to $1.95 billion in option fees and milestone payments. The deals with Gilgamesh, which is also developing treatments for anxiety and post-traumatic stress disorder, also launch AbbVie into the race to develop psychedelic compounds for psychiatric conditions -- a potential $50 billion market, according to Cantor Fitzgerald analyst Josh Schimmer. The deal, which includes an upfront payment and development milestones, could also bolster AbbVie's neurological conditions portfolio after its experimental schizophrenia drug, which it gained access to through an $8.7 billion purchase of Cerevel Therapeutics, failed in two mid-stage studies last year. Gilgamesh's lead candidate for depression, bretisilocin, activates the 5-HT2A serotonin receptor -- also targeted by classic psychedelics such as psilocybin, found in magic mushrooms, and LSD. The companies said bretisilocin has been shown to exert a shorter duration of psychoactive experience while retaining an extended therapeutic benefit in early and mid-stage studies. AbbVie will advance the drug into late-stage studies. "Large Pharma has been less active exploring psychedelic compounds due to potential regulatory concerns ... making today's deal more significant," said BMO Capital Markets analyst Evan Seigerman.

Read more of this story at Slashdot.

Amazon is facing a proposed class action lawsuit alleging it misleads customers by advertising digital movies and TV shows as "purchases," when in reality buyers only receive revocable licenses that can disappear if Amazon loses distribution rights. From the Hollywood Reporter: On Friday, a proposed class action was filed in Washington federal court against Amazon over a "bait and switch" in which the company allegedly misleads consumers into believing they've purchased content when they're only getting a license to watch, which can be revoked at any time. [...] The lawsuit accuses Amazon, which didn't respond to a request for comment, of misrepresenting the nature of movie and TV transactions during the purchase process. On its website and platform, the company tells consumers they can "buy" a movie. But hidden in a footnote on the confirmation page is fine print that says, "You receive a license to the video and you agree to our terms," the complaint says. The issue is already before a court. In a 2020 lawsuit alleging unfair competition and false advertising over the practice, Amazon maintained that its use of the word "buy" for digital content isn't deceptive because consumers understand their purchases are subject to licenses. Quoting Webster's Dictionary, it said that the term means "rights to the use or services of payment" rather than perpetual ownership and that its disclosures properly warn people that they may lose access. The court ultimately rebuffed Amazon's bid to dismiss the lawsuit outside of a claim alleging a violation of Washington's unjust enrichment law.

Read more of this story at Slashdot.

Russia's new Soyuz-5 rocket is set for a December debut as Moscow seeks to end reliance on Ukrainian technology and replace its aging Proton-M fleet. Ars Technica reports: According to the report, translated for Ars by Rob Mitchell, the debut launch of Soyuz-5 will mark the first of several demonstration flights, with full operational service not expected to begin until 2028. It will launch from the Baikonur spaceport in Kazakhstan. From an innovation standpoint, the Soyuz-5 vehicle does not stand out. It has been a decade in the making and is fully expendable, unlike a lot of newer medium-lift rockets coming online in the next several years. However, for Russia, this is an important advancement because it seeks to break some of the country's dependency on Ukraine for launch technology.

Read more of this story at Slashdot.

Two of Japan's largest media groups are suing AI search engine Perplexity over alleged copyright infringement, joining a growing list of news publishers taking legal action against AI companies using their content. FT: Japanese media group Nikkei, which owns the Financial Times, and the Asahi Shimbun newspaper said in statements on Tuesday that they had jointly filed a lawsuit in Tokyo. The groups join a number of Western media companies taking legal action against Perplexity, which provides answers to questions with sources and citations, using large language models (LLMs) from platforms such as OpenAI and Anthropic. The Japanese news providers claim Perplexity has, without permission, "copied and stored article content from the servers of Nikkei and Asahi" and ignored a "technical measure" designed to prevent this from happening. They claim that Perplexity's answers have given incorrect information attributed to the newspapers' articles, which "severely damages the credibility of newspaper companies."

Read more of this story at Slashdot.

An anonymous reader quotes a report from Ars Technica: Greenhouse gas reduction is no longer a priority for the US government, but if you're looking for a new vehicle and want to buy something with the lowest life cycle carbon emissions, you're best off looking for a compact with a small battery. That's one of the findings from a group at the University of Michigan of a comprehensive study that calculates the overall cradle-to-grave carbon impact for different types of vehicles, including factors like powertrain options, location (within the country), and use patterns. Even better, they built a tool you can use yourself. The study, published in Environmental Science and Technology, compares internal combustion engine powertrains with hybrid, 35- and 50-mile range plug-in hybrids, and 200-mile, 300-mile, and 400-mile battery electric powertrains across compact and midsize sedans, small and midsize SUVs, and pickup trucks, using a life cycle assessment model developed by Argonne National Laboratory and data of model year 2025 vehicles from the Environmental Protection Agency. If you expected that a gas-powered pickup truck would have the biggest carbon footprint, you'd be right. With a driving profile of 43 percent city driving and the rest highways (no cargo), a pickup will emit about 486 g CO2e per mile. Compared to that, a compact electric sedan with a 200-mile battery has just 17 percent of the life cycle emissions and is responsible for just 81 g CO2e per mile. A short-range electric pickup -- maybe that Slate that so many are salivating over -- is nearly as good, with a footprint that's only 25 percent the size of the gas pickup truck. On the other hand, hybrid powertrains (the kind that don't plug in) only reduce life cycle carbon compared to internal combustion alone by a modest amount -- between 11 and 13 percent, depending on the vehicle class. Plug-in hybrids with 35 miles of range can reduce emissions compared to plain combustion by 53-56 percent; with 50-mile batteries the reduction is 56-60 percent, assuming the PHEVs were driven in electric mode for 58 percent and 69 percent of the time, respectively. When it comes to BEVs, the smallest battery pack always has the least environmental impact. BEV powertrains with 400 miles of range have lifecycle emissions that are 67-69 percent lower than an ICE powertrain in the same vehicle. For 300-mile BEVs, this is an 81-83 percent reduction. A 200-mile BEV can be expected to contribute just 25-26 percent as much CO2e as an equivalent gas-burning vehicle would. That's not because EVs with big batteries are inefficient -- far from it -- but because making a battery for an EV is a very energy-intensive process. Most emissions from internal combustion engine (92 percent) and hybrid (89 percent) vehicles come from their use on the roads. But this changes once you start adding significant kWh-worth of battery. For PHEVs, the use phase is more like 73-80 percent, and for BEVs, it's just 48-60 percent, depending on the size of the batteries. The researchers also modeled different driving behaviors, including the use cases of someone who uses their vehicle just to commute and run errands; the "occasional road-tripper," most of whose needs are met by a small battery; and a contractor or someone else who has to drive a lot for work, with varying amounts of cargo onboard. As we've known for some time, where you get your energy from affects how clean your EV will be, and switching from gasoline to an EV has more of an impact in Seattle (which relies on hydropower) versus Cincinnati (where the electricity comes from burning coal), for both PHEVs and BEVs.

Read more of this story at Slashdot.

Farmers Insurance disclosed a breach affecting 1.1 million customers after attackers exploited Salesforce in a widespread campaign involving ShinyHunters and allied groups. According to BleepingComputer, the hackers stole personal data such as names, birth dates, driver's license numbers, and partial Social Security numbers. From the report: The company disclosed the data breach in an advisory on its website, saying that its database at a third-party vendor was breached on May 29, 2025. "On May 30, 2025, one of Farmers' third-party vendors alerted Farmers to suspicious activity involving an unauthorized actor accessing one of the vendor's databases containing Farmers customer information (the "Incident")," reads the data breach notification (PDF) on its website. "The third-party vendor had monitoring tools in place, which allowed the vendor to quickly detect the activity and take appropriate containment measures, including blocking the unauthorized actor. After learning of the activity, Farmers immediately launched a comprehensive investigation to determine the nature and scope of the Incident and notified appropriate law enforcement authorities." The company says that its investigation determined that customers' names, addresses, dates of birth, driver's license numbers, and/or last four digits of Social Security numbers were stolen during the breach. Farmers began sending data breach notifications to impacted individuals on August 22, with a sample notification [1, 2] shared with the Maine Attorney General's Office, stating that a combined total of 1,111,386 customers were impacted. While Farmers did not disclose the name of the third-party vendor, BleepingComputer has learned that the data was stolen in the widespread Salesforce data theft attacks that have impacted numerous organizations this year. Further reading: Google Suffers Data Breach in Ongoing Salesforce Data Theft Attacks

Read more of this story at Slashdot.

Mark Tyson writes via Tom's Hardware: On this day 34 years ago, an unknown computer science student from Finland announced that a new free operating system project was "starting to get ready." Linus Benedict Torvalds elaborated by explaining that the OS was "just a hobby, [it] won't be big and professional like GNU." Of course, this was the first public outing for the colossal collaborative project that is now known as Linux. Above, you can see Torvalds' first posting regarding Linux to the comp.os.minix newsgroup. The now famously caustic, cantankerous, curmudgeon seemed relatively mild, meek, and malleable in this historic Linux milestone posting. Torvalds asked the Minix community about their thoughts on a free new OS being prepared for Intel 386 and 486 clones. He explained that he'd been brewing the project since April (a few months prior), and asked for direction. Specifically, he sought input about other Minix users' likes and dislikes of that OS, in order to differentiate Linux. The now renowned developer then provided a rough summary of the development so far. Some features of Linux that Torvalds thought were important, or that he was particularly proud of, were then highlighted in the newsgroup posting. For example, the Linux chief mentioned his OS's multithreaded file system, and its absence of any Minix code. However, he humbly admitted the code as it stood was Intel x86 specific, and thus "is not portable." Last but not least, Torvalds let it be known that version 0.01 of this free OS would be out in the coming month (September 1991). It was indeed released on September 17, 1991, but someone else decided on the OS name at the last minute. Apparently, Torvalds didn't want to release his new OS under the name of Linux, as it would be too egotistical, too self-aggrandizing. He preferred Freax, a portmanteau word formed from Free-and-X. However, one of Torvald's colleagues, who was the administrator for the project's FTP server, did not think that 'Freax' was an appealing name for the OS. So this co-worker went ahead and uploaded the OS as 'Linux' on that date in September, without asking Torvalds.

Read more of this story at Slashdot.

Struggling small biotech firms are pivoting into cryptocurrencies, rebranding as "crypto treasuries" or stockpiling digital assets like Ether and Litecoin as a last-ditch effort to boost share prices amid stalled funding and weak drug pipelines. Bloomberg reports: Shares of 180 Life Sciences Corp., now doing business as ETHZilla, tripled after the Peter Thiel-backed company said it had accumulated Ether tokens worth over $350 million. Less than two weeks later, the stock's gains have been erased. In July, Sonnet BioTherapeutics Holdings soared 243% in one volatile session on plans to transform into a public crypto treasury while MEI Pharma Inc. initially doubled on plans to sell shares to fund a Litecoin treasury. Such about-faces are a tried-and-true formula for small firms when funds are low and shares are under pressure. For drugmakers it can be a sudden shift to chase after trendy new treatment targets, still other companies rebrand with buzzwords like artificial intelligence to juice returns. Now some biotech executives are using digital coins to pump new life into flagging shares. So far in 2025, at least 10 biotechs have announced a pivot into digital assets. The announcements frequently spark frenzied, but short-lived, spikes in shares. "If they're low on ideas, if they can't find relevance in drug development, they're going to try to justify their existence as management in another way," according to Mike Taylor, lead portfolio manager of the Simplify Health Care ETF. "You have a handful of companies trying to reinvent themselves into some other tangent. And, most, if not all won't work out."

Read more of this story at Slashdot.

An anonymous reader quotes a report from Ars Technica: AMD's X3D-series Ryzen chips have become popular with PC gamers because games in particular happen to benefit disproportionately from the chips' extra 64MB of L3 cache memory. But that extra memory occasionally comes with extra headaches. Not long after they were released earlier this year, some early adopters started having problems with their CPUs, ranging from failure to boot to actual physical scorching and burnout -- the problems were particularly common for users of the 9800X3D processor in ASRock motherboards, and one Reddit thread currently records 157 incidents of failure for that CPU model across various ASRock boards. In an interview with the Korean language website Quasar Zone (via Tom's Hardware), AMD executives David McAfee and Travis Kirsch acknowledged the problems and pointed to the most likely culprit: motherboard makers who don't follow AMD's recommended specifications. Some manufacturers have historically shipped their AMD and Intel motherboards with elevated default power settings in the interest of squeezing a bit more performance out of the chips -- but those adjustments can also cause problems in some cases, especially for higher-end CPUs.

Read more of this story at Slashdot.

Apple has filed a lawsuit against former Apple Watch staffer Dr. Chen Shi, alleging that he "conspired to steal Apple's trade secrets relating to Apple Watch and to disclose them to his new employers (Oppo)." The company alleges he downloaded 63 sensitive documents, attended technical meetings, and coordinated with Oppo to transfer proprietary information, though Oppo denies wrongdoing. The Verge reports: Ahead of starting his new job at Oppo, the employee, Dr. Chen Shi, attended "dozens" of meetings with technical members on the Apple Watch team to learn about their work and downloaded 63 documents "from a protected Box folder" that he loaded onto a USB drive, according to the lawsuit. Shi allegedly sent a message to Oppo saying that he was working to "collect as much information as possible" before starting his job. And he searched the internet for terms like "how to wipe out macbook" and "Can somebody see if I've opened a file on a shared drive?" from his Apple-issued MacBook before leaving the company. Shi was formerly a sensor system architect at Apple, and the company says he had "a front row seat to Apple's development of its cutting-edge health sensor technology, including highly confidential roadmaps, design and development documents, and specifications for ECG sensor technology." He now heads up a team working on sensing technology at Oppo -- which Apple says it learned because of "messages he left on his Apple-issued work iPhone." In his resignation letter to Apple, Shi said he was leaving "due to personal and family reasons." Via that iPhone, Apple also says it found messages from Oppo demonstrating that it "encouraged, approved, and agreed to Dr. Shi's plan to collect Apple's proprietary information before leaving Apple."

Read more of this story at Slashdot.