Linux fréttir

alternative_right writes: A study by Anthropic researchers on how human feedback can encourage sycophantic behavior showed that AI assistants will sometimes modify accurate answers when questioned by the user -- and ultimately give an inaccurate response.

Read more of this story at Slashdot.

Apple displayed a full-screen ad for "F1 The Movie" in its TV app that linked directly to a web browser for ticket purchases without showing warning screens that the company requires other developers to include when directing users outside their apps. The "Buy Tickets" button sent users to the F1 movie website in their default browser without confirmation dialogs or interstitial warnings. Apple mandates that third-party developers show scare sheets when linking out of apps to sell digital content, but considers movie tickets a "real-world experience" exempt from its In-App Purchase system. Further reading: iPhone Customers Upset By Apple Wallet Ad Pushing F1 Movie.

Read more of this story at Slashdot.

The iMaker's fight with European regulators continues

Apple is on the hook for a €500 million (US $587 million) anti-steering fine in the EU, so it's reportedly doing what any profit-driven enterprise in such a position would do: Appealing.…

An anonymous reader shares a report: Netflix doubled down on its global anime strategy over the weekend, unveiling a slate of new titles and fresh footage during its showcase at Anime Expo in Los Angeles. The company also shared updated viewership data highlighting just how far Japanese anime has come in expanding from its former niche into a powerhouse global content category. According to Netflix, more than 50 percent of its members -- amounting to over 150 million households, or an estimated 300 million viewers -- now watch anime. The company says anime viewership on the platform has tripled over the past five years, with 2024 marking a record-breaking year: 33 anime titles appeared in Netflix's Global Top 10 (Non-English) rankings, more than double the number in 2021.

Read more of this story at Slashdot.

An anonymous reader shares a report: OpenAI said it has no active plans to use Google's in-house chip to power its products, two days after Reuters and other news outlets reported on the AI lab's move to turn to its competitor's artificial intelligence chips to meet growing demand. A spokesperson for OpenAI said on Sunday that while the AI lab is in early testing with some of Google's tensor processing units (TPUs), it has no plans to deploy them at scale right now.

Read more of this story at Slashdot.

The European Union is holding back on signing a joint climate action pledge with China at a summit this month to mark a half-century of diplomatic ties, a top climate official told the Financial Times in remarks published on Monday. Reuters: The EU's climate targets are among the world's most ambitious, but they have been based entirely on domestic emissions cuts. Now the bloc faces a mid-September deadline to submit a new 2035 climate target to the United Nations. Brussels has refused Beijing's repeated requests for a mutual climate commitment after the summit of the world's second- and third-largest economies, unless China promises to do more to cut greenhouse gas emissions, EU officials said. "There is only merit in having a declaration from our perspective if there are also content nuts to be cracked and ambition to be displayed," Climate Commissioner Wopke Hoekstra told the paper.

Read more of this story at Slashdot.

Americans live in separate economic realities: Those with a job are likely to stay employed, but those without one are likely to stay unemployed. From a report: Welcome to the low-hire, low-fire labor market. Private-sector layoffs are at historic lows, but that masks a dreadful outlook for unemployed workers or those unhappy with their current positions. [...] "We're in a complex jobs market -- it's not falling apart but the lack of dynamism, the lack of churn and the lack of hiring has been punctuated in the first half of the year," says ADP chief economist Nela Richardson. "Many employers are loath to lay off workers until they see the whites of the eyes of a recession, having had such problems finding suitable workers in the first place," David Kelly, chief global strategist at J.P. Morgan Asset Management, wrote in a recent note.

Read more of this story at Slashdot.

Poland generated more electricity from renewables than coal for the first time in June, marking a key moment in the country's efforts to cut its reliance on the most polluting fossil fuel. From a report: The shift comes as Prime Minister Donald Tusk's government accelerates efforts to diversify energy production in Poland, which despite recent progress remains a major producer of coal and the most coal-dependent country in the EU, with about 60 per cent of its electricity coming from the fossil fuel in 2024. Last month renewable energy sources accounted for 44.1 per cent of Poland's electricity mix, narrowly surpassing coal, which fell to 43.7 per cent, according to a study to be published next Monday by Forum Energii, a Warsaw-based energy think-tank, using data from Poland's grid operator. Natural gas made up the remainder.

Read more of this story at Slashdot.

Lucy in the sky with calcium

Astroboffins have found the first evidence of a double-detonated Type Ia supernova, which could explain why we have enough bright points of reference in the skies to plot our place in the universe.…

Springer Nature published a $169 machine learning textbook in April containing citations that appear to be largely fabricated, according to an investigation by Retraction Watch. The site checked 18 of the 46 citations in "Mastering Machine Learning: From Basics to Advanced" by Govindakumar Madhavan and found two-thirds either did not exist or contained substantial errors. Three researchers contacted by Retraction Watch confirmed their supposedly authored works were fake or incorrectly cited. Yehuda Dar of Ben-Gurion University said a paper cited as appearing in IEEE Signal Processing Magazine was actually an unpublished arXiv preprint. Aaron Courville of Universite de Montreal confirmed he was cited for sections of his "Deep Learning" book that "doesn't seem to exist." The pattern of nonexistent citations matches known hallmarks of large language model-generated text. Madhavan did not answer whether he used AI to generate the book's content. The book contains no AI disclosure despite Springer Nature policies requiring authors to declare AI use beyond basic copy editing.

Read more of this story at Slashdot.

As power concerns beset builds, this floating datacenter can plug into powership next door

Japanese shipping biz Mitsui OSK Lines (MOL) is planning to fit out a ship as a floating datacenter that can draw energy from the shore or from an accompanying powership.…

Line-judging tech flubs crucial point, leaving players and fans seeing red

"You cannot be serious" was likely uttered by more than a few folk watching Russia's Anastasia Pavlyuchenkova versus Britain's Sonay Kartal at Wimbledon yesterday after the tennis tournament's AI line-calling tech dropped the ball.…

Official notice confirms delay to cutoff until the end of July. Not to worry, AI modelling's in the wings

The US defense department satellite service that's cutting off the flow of data used for hurricane forecasting is doing so "to mitigate a significant cybersecurity risk" to government "high performance computing environments."…

Get your creds in order or risk BEC, ransomware attacks, orgs warned

A rise in advanced phishing kits and info-stealing malware are to blame for a 156 percent jump in cyberattacks targeting user logins, say researchers.…

An anonymous reader shares a report: India is set to begin mass-producing electric-vehicle batteries within 18 months, a step hailed as a leap towards industrial self-reliance. Yet the structure of this new industry looks troublingly familiar, echoing a pattern of dependence that has long marked India's economy. Nowhere is this dependence clearer than in the heft of intellectual property. The portfolios of India's largest battery-makers, Amara Raja and Exide, contain just seven patents combined. This pales in comparison to the industry's giants: China's CATL sits on a hoard of over 43,000 patents, while South Korea's LG Energy Solution possesses some 70,000. Having largely missed the global lithium-ion boom, India's established lead-acid manufacturers built a business model on licensing technology rather than inventing it. This long-standing habit is now reflected in deals that create deep technological dependency. A 2022 agreement between Exide and China's SVOLT, for example, calls for SVOLT to not only transfer intellectual property but also to oversee plant construction, supply the equipment and integrate the factory into its own Chinese supply chain. Amara Raja's deal with Gotion High-Tech in June 2024 follows a similar template.

Read more of this story at Slashdot.

China "is pouring money into building an AI supply chain with as little reliance on the U.S. as possible," the Wall Street Journal noted this weekend. But what does that look like? The New York Times visits Liangzhu, "the coder 'village' at the heart of China's AI frenzy... a quiet suburb of the southern Chinese city of Hangzhou... As China faces off with the United States over tech primacy, Hangzhou has become the centre of China's AI frenzy," with its proximity to tech companies like Alibaba and DeepSeek..." In Liangzhu, many engineers said they were killing time until they could create their own startups, waiting out noncompete agreements they had signed at bigger companies like ByteDance... But some said the government support for Hangzhou's tech scene had scared off some investors. Several company founders, who asked not to be named so they could discuss sensitive topics, said it was difficult for them to attract funds from foreign venture capital firms, frustrating their ambitions to grow outside China. The nightmare situation, they said, would be to end up like ByteDance, the Chinese parent of TikTok, whose executives have been questioned before Congress about the company's ties to the Chinese government. Founders described choosing between two paths for their companies' growth: Take government funding and tailor their product to the Chinese market, or raise enough money on their own to set up offices in a country like Singapore to pitch foreign investors. For most, the first was the only feasible option. Another uncertainty is access to the advanced computer chips that power artificial intelligence systems. Washington has spent years trying to prevent Chinese companies from buying these chips, and Chinese companies like Huawei and Semiconductor Manufacturing International Corp. are racing to produce their own. So far, the Chinese-made chips work well enough to help companies like ByteDance provide some of their AI services in China. Many Chinese companies have created stockpiles of Nvidia chips despite Washington's controls. But it is not clear how long that supply will last, or how quickly China's chipmakers can catch up to their American counterparts... Liangzhu villagers have been hosting film nights. They had recently gathered to watch "The Matrix." Afterward, they decided the movie should be required viewing, Lin said. Its theme — people finding their way out of a vast system controlling society — provided spot-on inspiration. Aspiring founders in Liangzhu, even those who did not go to top universities, believe they could start the next world-changing tech company, said Felix Tao [a 36-year-old former Facebook and Alibaba employee.] "Many of them are super brave to make a choice to explore their own way, because in China that is not the common way to live your life."

Read more of this story at Slashdot.

Digital map of subterranean infrastructure promised in 2021 set to launch by year end

Ordnance Survey, the UK's official map maker, is seeking a tech supplier to help it obtain and manage data from utilities companies for a project that aims to avoid damage to subterranean infrastructure, which costs around £2.4 billion a year.…

Tips on who pays when staff don't transfer, when the regulations apply ... and when they don't

Comment Few IT leaders or staffers realize just how much automation, AI, and cloud delivery are disrupting the legal and human frameworks that underpin outsourcing - especially when it comes to the Transfer of Undertakings (Protection of Employment) Regulations 2006, better known as TUPE.…

A virtual environment makes a great de-hype advisor

Opinion In human imagination, AIs have been good for two things: trying to take over, and loving a good game. The earliest post-war AI thinkers took it almost for granted that once computers could beat humans at chess, true artificial intelligence would have arrived. Such thinking was disproved 50 years on when IBM's Deep Blue beat Kasparov in 1997. Computers could be very, very good at chess while still having the IQ of a pebble.…

"Citizen scientists have successfully located thousands of previously unknown pairs of 'eclipsing binary' stars," reports the Washington Post, citing a recent announcement from NASA. The ongoing initiative helps space researchers hunt for "eclipsing binary" stars, a rare phenomenon in which two stars orbit one another, periodically blocking each other's light. These star pairs offer important data to astrophysicists, who consider the many measurable properties of eclipsing binaries — and the information they bear about the history of star formation and destruction — as a foundation of the field... The citizen science project in question, the Eclipsing Binary Patrol, validates images from NASA's Transiting Exoplanet Survey Satellite (TESS) mission. The satellite, launched in 2018, is "exceptionally capable at detecting varying stars," the researchers write in a preprint paper describing the initiative. The researchers used machine learning to identify about 1.2 million potential eclipsing star pairs. Citizen scientists then validated a subset of about 60,000... manually inspecting hundreds of thousands of images of eclipse-like events and weeding out actual binaries from images that tricked the algorithm. "Thankfully," the researchers write, "to the rescue come volunteers from all walks of life that boost the capacity of bandwidth-limited professional astronomers many-fold and help tackle the ever-increasing volume of publicly available astronomical data." Universe Today describes how they limited the dataset to only stars with a magnitude brighter than 15, then used a Python tool to generate a massive dataset of millions of light curves... The outcome of all the work resulted in the identification of 10,001 eclipsing binary systems. 7,936 of them are new to science, while the other 2,065 were previously known, but the study provided updated, more accurate, parameters for their periods, as TESS' dataset provided better insight. There were also some particularly interesting systems that could hold new discoveries, including several that had variable eclipse timings, and plenty that might have a third star, and some that show a significant dynamic between the star being orbited and the one doing the orbiting. All of those systems await further research, but there's another, unspoken factor at play in this data — exoplanets. TESS was originally designed as an exoplanet hunter, and this kind of large scale AI/human collaboration of lightcurve analysis is exactly the kind of work that could potentially produce even more accurate exoplanet catalogues, as evidenced by some of the work already done in this paper. That seems to be the next step for this dataset, with Dr. Kostov telling an interviewer "I can't wait to search them for exoplanets!" Given the data has already been collected, and the team has already been assembled, it's very likely he'll get his chance soon.

Read more of this story at Slashdot.