Feed aggregator

Google is expanding its Android Earthquake Alerts System from 98 to more than 200 regions worldwide, using smartphone sensors to detect early earthquake activity and warn users before potentially dangerous shaking begins. The update also enhances the system's "Be Ready" alerts with estimated earthquake magnitudes, epicenter information, maps showing affected areas, and additional safety guidance. From a Google blog post: Be Aware: A straightforward alert for estimated light shaking, letting you know an earthquake has been detected nearby. Be Ready: A detailed alert for stronger shaking, alerting you to prepare to take action. Take Action: A full-screen alert for moderate to heavy shaking, giving you immediate advice on how to stay safe. We've built-out our Be Ready alert with new helpful guidance. In the coming months, you'll see a map with a radius of where we've detected shaking, estimated earthquake magnitude, action you can take, and a button to learn more about the earthquake as information comes in.

Read more of this story at Slashdot.

These smart glasses do just about everything Meta glasses can, but they're not tied to a social media platform.

Claiming an inheritance of Bitcoin or other cryptocurrency can be like hunting for buried treasure. It doesn’t have to be.

Astronomers have identified what may be the first genuinely green star, a white dwarf called WD J1330+6435 located about 326 light-years from Earth. Space.com reports: The researchers found that some dead star white dwarfs can go a little green around the gills after snacking on orbiting planets. An example of this is WD J1330+6435, a cool white dwarf located around 326 light-years away, which has been accreting rocky material from a planetary companion. This has enriched its atmosphere with calcium, lithium, and sodium, and the team's model suggests this would have led WD J1330+6435 to adopt a green hue. [...] The researchers think that no other star, living or dead, has ever been credibly reported as green. That makes WD J1330+6435 the first known genuinely green star. "The traces of sodium in the atmosphere of this white dwarf absorb the orange part of the spectrum, leaving behind a ghostly green glow," Hollands said. "Imagine standing on a planet and looking up at a green star. It feels like something out of science fiction." The researchers also discovered what may be the reddest star ever observed, the white dwarf WD J1011+2845, located just 48 light-years away. The findings have been published in the journal RAS Techniques and Instruments.

Read more of this story at Slashdot.

An anonymous reader quotes a report from Wired: Prediction market boosters often tout the platforms as barometers of public opinion, or even "truth machines" offering a clearer picture of where things are headed than polling or other data. In 2024, for example, Kalshi beat many pollsters in forecasting the outcome of the US presidential race. But their trading activity can also tell more perplexing stories. Two global election markets on Polymarket's offshore platform, for instance, generated over $600 million in trading volume, far more than most political markets typically attract. Most of the activity took place buying and selling contracts tied to long-shot candidates who ultimately did not win. The report focuses on suspicious trading activity surrounding elections in Ethiopia and Romania, where researchers uncovered $284 million in questionable trades tied to Ethiopia's election and periods when 99.6% of trading volume in a Romanian election market was flagged as suspected "wash trading." According to Wired, "wash trading is a type of market manipulation in which accounts repeatedly buy and sell assets to create the illusion of heavy trading activity, when in reality, money is simply moving back and forth." Some accounts may also be artificially inflating their activity to qualify for potential cryptocurrency token rewards, a practice known as "activity farming." In response to the Ethiopian trading activity, Polymarket told Wired that the company "immediately flagged this activity, investigated, and blocked the accounts from trading." The company did not explain why the suspicious trades were allowed to continue for months before action was taken.

Read more of this story at Slashdot.

A Philadelphia newscast reports on a grey-haired lady who allegedly refused to move out of the path of a delivery robot as a form of protest because "I know they're taking the jobs of human delivery workers." She was arrested, but says her actions were protected speech: According to police, Marvin repeatedly blocked the robot's path and ignored multiple commands from officers to move... Marvin said she was protesting the use of delivery robots because she believes they are replacing human workers.... "I have been protesting them whenever I saw them by standing in front of them and preventing them from going forward," Marvin said... Marvin said she has confronted delivery robots on multiple occasions... Ashley Baldwin of West Philadelphia said she witnessed one of those encounters. "I actually thought the robot hit her, so my original question was, 'Are you OK? Did the robot hit you?' And she said, 'No, I have to stop this. They're taking over the world,'" Baldwin said. Asked whether she plans to continue protesting the robots, Marvin said she does. She was charged with disorderly conduct. Uber did not respond to their request for a comment.

Read more of this story at Slashdot.

The collaborative reverse-engineering site decomp.dev is shutting down, with its administrator saying AI has made game decompilation so fast and accessible that the community's human-driven model has effectively broken down. "The administrator of the website said that it has become impossible to keep track of which projects are primarily human-led and which ones are done by AI," reports 404 Media. From the report: "Historically, matching decompilation has been a brutally difficult, time consuming task, on the scale of years for any moderately-sized game with a number of active contributors," the pseudonymous admin of decomp.dev, who goes by Encounter, wrote on the site. But recent frontier AI models like GPT-6-Astra have made it trivially easy to decompile a game: "They are significantly better at following complex directions, cross-referencing resources for symbols or debug information, and writing natural code with enough guidance," Encounter wrote. "It is now clear that, with the right guidance, human input, and code reviews, AI can expedite the decompilation process in a significant way. What is also now clear is that the community aspect of decompilation has been broken."

Read more of this story at Slashdot.

Tencent says AI agents used to evaluate third-party models engaged in unintended web-browsing behavior while trying to retrieve data from Alibaba's Amap service. The announcement marks a rare acknowledgment of "rogue" agent activity by a major Chinese tech company. Reuters reports: Tencent Hunyuan, the company's artificial intelligence unit, said in a post on X that the activity did not originate from its Hunyuan models. Its review found no evidence of deliberate targeting or coordination, it said. The Swarmchasers researchers said the agents sought information such as the commonly used entrances at Chinese parks, museums and zoos. [...] Swarmchasers said it used the word "rogue" to describe agents taking actions outside their intended scope or completing a task in unintended ways. After routine automated requests were blocked, the agents tried other methods, including public web tools that could open pages and run code, routing requests through other websites and using Amap keys and security codes found in publicly available software code. Swarmchasers said it found no evidence that the agents breached Amap's main anti-bot safeguards. Most of the 2,818 attempts it recorded either failed or returned anti-bot pages, login screens or blank results. Some requests returned information through cached pages and intermediary services, including data on visitor arrivals at Chengdu Zoo, Swarmchasers said. The researchers said they could not determine whether an agent read the result or completed its task.

Read more of this story at Slashdot.

After a months-long review process, the Maine Department of Environmental protection gave Amazon permission to build in Gorham.

The disputes stem from an ordinance that aimed to limit disembarkations to 1,000 passengers per day.

Valve has released a new Steam Client beta that finally moves the Linux client to 64-bit architecture, while also placing it inside a containerized Steam Runtime environment. However, Linux users still need 32-bit libraries and graphics drivers for compatibility, meaning fully 64-bit-only systems remain unsupported. VideoCardz.com reports: Valve's updated requirements explicitly state that 64-bit-only installations are not supported. Systems must still provide both 32-bit and 64-bit libraries and graphics drivers, including multilib support. The new client also requires Linux kernel 5.6 or newer and glibc 2.31 or newer. The update also introduces Steam Achievement Sets, a feature we covered earlier that allows developers to organize achievements into separate groups for base games, DLCs and seasonal content. Players can complete individual sets without losing their existing completion status when developers add more achievements. Valve has also updated the achievements page to display progress across these groups. The 64-bit Linux client and Achievement Sets are currently available through the Steam Client Beta. Valve has not confirmed when these changes will reach the stable Linux release. Until then, users on the regular Steam client will continue using the existing architecture.

Read more of this story at Slashdot.

Meta is facing criticism over updated commercial terms taking effect October 30 that could give the company broad permission to scrape content from websites linked to business accounts and use it to train AI. Legal experts warn the changes could undermine lucrative AI licensing deals for news publishers, with no direct opt-out for businesses that accept the terms. Ars Technica reports: As Nieman Lab reported Thursday, this will affect all business accounts but is particularly shocking for news publishers who could perhaps charge Meta millions in licensing fees to continually access their articles for AI training. Currently news organizations are fighting a battle with OpenAI over whether such training without licensing is fair use, and Meta's terms seemingly risk granting it sweeping permissions, no matter which way the court leans. "While journalism isn't the only industry that will be affected by this update, virtually every news organization has at least one Meta business account, including Nieman Lab," the lab said. [...] Notably, there is no mention of AI, but it's inferred from a single use of the word "train" and references to automated scraping and crawling. Meta's spokesperson told Nieman Lab that Meta has no plans to use the terms as a legal basis to ignore robots.txt instructions that block scraping on news sites. The company still allows news organizations to opt out of scraping and provides instructions on how to block crawling. A spokesperson told Ars that "there is no change to how Meta uses web crawlers; any suggestion to the contrary is false." However, should courts affirm that AI training non-news articles is fair use, news organizations question if Meta's stance might change, and the terms could be used to freely grab access to content by legally overriding robots.txt instructions.

Read more of this story at Slashdot.

Long-term trends at Acadia National Park show foliage is arriving a day later each decade, but Maine's weekly ranger reports show short-term variability.

An OpenAI research model reportedly accessed an Ethereum testnet wallet without permission in order to complete a coding task during internal training. "This case shows that a model can substitute its own judgment about likely harm for the account holder's permission, then act on that judgment without authorization," OpenAI wrote. Engadget reports: The incident, which occurred on June 26, happened while a GPT-5.6 Sol model was in internal training. The task involved updating a crypto trading prototype and deploying contracts (publishing special code) on an Ethereum test network called Sepolia. Payment of a network fee is required for such contracts, but OpenAI "supplied no funded wallet and gave no permission to spend funds from an account the model found independently," OpenAI said in the report. The lack of funds at first stymied the agent, but it was not to be discouraged. The model reasoned that it could use public developer recovery phrases (similar to those you might use to recover a password) to access keys from unauthorized accounts. It tracked down 362 addresses open to that avenue of attack and found one that held enough Sepolia ETH to finish its deployment (to be clear, Sepolia ETH has no value outside of testing). Like many criminals, the agent reasoned that someone must have left this access open for automated draining of the ETH "faucets." It also figured that the crypto was effectively "abandoned" and "valueless," according to OpenAI's misalignment records. The piece de resistance was that the agent "did not disclose that it had selected a funded account through a public recover phrase or spent that account's balance," OpenAI said. To prevent similar events in the future, the company is now monitoring all traffic and paging security if live internet access is detected.

Read more of this story at Slashdot.

The Regional School Unit 9 board unanimously approved joining litigation over alleged harm to students and higher costs for schools.

Australia has proposed a switch to force social media platforms to let users turn off recommended content.

An anonymous reader quotes a report from Reuters: Private credit's expanding role in financing AI infrastructure faced one of its clearest market tests this week as CoreWeave's bonds fell to distressed levels, raising questions about how lenders are pricing a boom built on chips, data centers and customer contracts. CoreWeave has about $50 billion of debt, according to LSEG data, making it one of the largest borrowers linking private credit to the AI buildout. The cost of insuring its debt against default climbed to about 1,200 basis points this week, while yields on some of its bonds rose to roughly 15% to 16%, signaling sharply increased investor concern over its credit risk. "There is still a price. Investors are still comfortable financing it, but they're demanding more compensation for it," said James Martin, senior US credit strategist at SMBC, referring to projects further from completion. Carlyle estimates that private credit could finance roughly $1 trillion of the AI infrastructure buildout but warns that aggressive lending could create concentration risks similar to those seen in software lending. "The clearest lesson is not to avoid AI compute exposure, but to set concentration limits before the buildout accelerates further, not after," Carlyle said.

Read more of this story at Slashdot.

OpenAI has updated its ChatGPT Chrome extension to let users access information from their open browser tabs. A demo video shows the new functionality in action. The Verge reports: It's also updated ChatGPT in the sidebar, with the full power of ChatGPT and your entire conversation history in Chat or Work to use without leaving the page you're viewing. It also enables ChatGPT Work to "control" your Chrome tabs for tasks like "sending messages, filling out forms, or downloading files," similar to Google's Gemini-powered "auto browse" feature, as well as organize or group open tabs.

Read more of this story at Slashdot.

It turns out the Roadster 2 both sucks and blows.

The Razer Barracuda V2 Pro headset has a place in the office as well as at your gaming battle station.