This month saw results from a yearlong global study of "potential negative risks that generative AI poses to student". The study (by the Brookings Institution's Center for Universal Education) also suggests how to prevent risks and maximize benefits: After interviews, focus groups, and consultations with over 500 students, teachers, parents, education leaders, and technologists across 50 countries, a close review of over 400 studies, and a Delphi panel, we find that at this point in its trajectory, the risks of utilizing generative AI in children's education overshadow its benefits. "At the top of Brookings' list of risks is the negative effect AI can have on children's cognitive growth," reports NPR — "how they learn new skills and perceive and solve problems." The report describes a kind of doom loop of AI dependence, where students increasingly off-load their own thinking onto the technology, leading to the kind of cognitive decline or atrophy more commonly associated with aging brains... As one student told the researchers, "It's easy. You don't need to (use) your brain." The report offers a surfeit of evidence to suggest that students who use generative AI are already seeing declines in content knowledge, critical thinking and even creativity. And this could have enormous consequences if these young people grow into adults without learning to think critically... Survey responses revealed deep concern that use of AI, particularly chatbots, "is undermining students' emotional well-being, including their ability to form relationships, recover from setbacks, and maintain mental health," the report says. One of the many problems with kids' overuse of AI is that the technology is inherently sycophantic — it has been designed to reinforce users' beliefs... Winthrop offers an example of a child interacting with a chatbot, "complaining about your parents and saying, 'They want me to wash the dishes — this is so annoying. I hate my parents.' The chatbot will likely say, 'You're right. You're misunderstood. I'm so sorry. I understand you.' Versus a friend who would say, 'Dude, I wash the dishes all the time in my house. I don't know what you're complaining about. That's normal.' That right there is the problem." AI did have some advantages, the article points out: The report says another benefit of AI is that it allows teachers to automate some tasks: "generating parent emails ... translating materials, creating worksheets, rubrics, quizzes, and lesson plans" — and more. The report cites multiple research studies that found important time-saving benefits for teachers, including one U.S. study that found that teachers who use AI save an average of nearly six hours a week and about six weeks over the course of a full school year... AI can also help make classrooms more accessible for students with a wide range of learning disabilities, including dyslexia. But "AI can massively increase existing divides" too, [warns Rebecca Winthrop, one of the report's authors and a senior fellow at Brookings]. That's because the free AI tools that are most accessible to students and schools can also be the least reliable and least factually accurate... "[T]his is the first time in ed-tech history that schools will have to pay more for more accurate information. And that really hurts schools without a lot of resources." The report calls for more research — and make several recommendations (including "holistic" learning and "AI tools that teach, not tell.") But this may be their most important recommendation. "Provide a clear vision for ethical AI use that centers human agency..." "We find that AI has the potential to benefit or hinder students, depending on how it is used."

Read more of this story at Slashdot.

An anonymous reader shared this report from the Guardian: Google's search feature AI Overviews cites YouTube more than any medical website when answering queries about health conditions, according to research that raises fresh questions about a tool seen by 2 billion people each month. The company has said its AI summaries, which appear at the top of search results and use generative AI to answer questions from users, are "reliable" and cite reputable medical sources such as the Centers for Disease Control and Prevention and the Mayo Clinic. However, a study that analysed responses to more than 50,000 health queries, captured using Google searches from Berlin, found the top cited source was YouTube. The video-sharing platform is the world's second most visited website, after Google itself, and is owned by Google. Researchers at SE Ranking, a search engine optimisation platform, found YouTube made up 4.43% of all AI Overview citations. No hospital network, government health portal, medical association or academic institution came close to that number, they said. "This matters because YouTube is not a medical publisher," the researchers wrote. "It is a general-purpose video platform...." In one case that experts said was "dangerous" and "alarming", Google provided bogus information about crucial liver function tests that could have left people with serious liver disease wrongly thinking they were healthy. The company later removed AI Overviews for some but not all medical searches... Hannah van Kolfschooten, a researcher specialising in AI, health and law at the University of Basel who was not involved with the research, said: "This study provides empirical evidence that the risks posed by AI Overviews for health are structural, not anecdotal. It becomes difficult for Google to argue that misleading or harmful health outputs are rare cases. "Instead, the findings show that these risks are embedded in the way AI Overviews are designed. In particular, the heavy reliance on YouTube rather than on public health authorities or medical institutions suggests that visibility and popularity, rather than medical reliability, is the central driver for health knowledge."

Read more of this story at Slashdot.

An anonymous reader shared this report from Fortune The large language models (LLMs) that have captivated the world are not a path to human-level intelligence, two AI experts asserted in separate remarks at Davos. Demis Hassabis, the Nobel Prize-winning CEO of Google DeepMind, and the executive who leads the development of Google's Gemini models, said today's AI systems, as impressive as they are, are "nowhere near" human-level artificial general intelligence, or AGI. [Though the artilcle notes that later Hassabis predicted there was a 50% chance AGI might be achieved within the decade.] Yann LeCun — an AI pioneer who won a Turing Award, computer science's most prestigious prize, for his work on neural networks — went further, saying that the LLMs that underpin all of the leading AI models will never be able to achieve humanlike intelligence and that a completely different approach is needed... ["The reason ... LLMs have been so successful is because language is easy," LeCun said later.] Their views differ starkly from the position asserted by top executives of Google's leading AI rivals, OpenAI and Anthropic, who assert that their AI models are about to rival human intelligence. Dario Amodei, the CEO of Anthropic, told an audience at Davos that AI models would replace the work of all software developers within a year and would reach "Nobel-level" scientific research in multiple fields within two years. He said 50% of white-collar jobs would disappear within five years. OpenAI CEO Sam Altman (who was not at Davos this year) has said we are already beginning to slip past human-level AGI toward "superintelligence," or AI that would be smarter than all humans combined... The debate over AGI may be somewhat academic for many business leaders. The more pressing question, says Cognizant CEO Ravi Kumar, is whether companies can capture the enormous value that AI already offers. According to Cognizant research released ahead of Davos, current AI technology could unlock approximately $4.5 trillion in U.S. labor productivity — if businesses can implement it effectively.

Read more of this story at Slashdot.

An anonymous reader shared this report from Reuters: U.S. insurer Lemonade said on Wednesday it would offer a 50% rate cut for drivers of Tesla electric vehicles when the automaker's Full Self-Driving (FSD) driver assistance software is steering because it had data showing it reduced accidents. Lemonade's move is an endorsement of Tesla CEO Elon Musk's claims that the company's vehicle technology is safer than human drivers, despite concerns flagged by regulators and safety experts. As part of a collaboration, Tesla is giving Lemonade access to vehicle telemetry data that will be used to distinguish between miles driven by FSD — which requires a human driver's supervision — and human driving, the New York-based insurer said. The price cut is for Lemonade's pay-per-mile insurance. "We're looking at this in extremely high resolution, where we see every minute, every second that you drive your car, your Tesla," Lemonade co-founder Shai Wininger told Reuters. "We get millions of signals emitted by that car into our systems. And based on that, we're pricing your rate." Wininger said data provided by Tesla combined with Lemonade's own insurance data showed that the use of FSD made driving about two times safer for the average driver. He did not provide details on the data Tesla shared but said no payments were involved in the deal between Lemonade and the EV maker for the data and the new offering... Wininger said the company would reduce rates further as Tesla releases FSD software updates that improve safety. "Traditional insurers treat a Tesla like any other car, and AI like any other driver," Wininger said. "But a driver who can see 360 degrees, never gets drowsy, and reacts in milliseconds isn't like any other driver."

Read more of this story at Slashdot.

TechCrunch reports: On Wednesday, Anthropic released a revised version of Claude's Constitution, a living document that provides a "holistic" explanation of the "context in which Claude operates and the kind of entity we would like Claude to be...." For years, Anthropic has sought to distinguish itself from its competitors via what it calls "Constitutional AI," a system whereby its chatbot, Claude, is trained using a specific set of ethical principles rather than human feedback... The 80-page document has four separate parts, which, according to Anthropic, represent the chatbot's "core values." Those values are: 1. Being "broadly safe." 2. Being "broadly ethical." 3. Being compliant with Anthropic's guidelines. 4. Being "genuinely helpful..." In the safety section, Anthropic notes that its chatbot has been designed to avoid the kinds of problems that have plagued other chatbots and, when evidence of mental health issues arises, direct the user to appropriate services... Anthropic's Constitution ends on a decidedly dramatic note, with its authors taking a fairly big swing and questioning whether the company's chatbot does, indeed, have consciousness. "Claude's moral status is deeply uncertain," the document states. "We believe that the moral status of AI models is a serious question worth considering. This view is not unique to us: some of the most eminent philosophers on the theory of mind take this question very seriously." Gizmodo reports: The company also said that it dedicated a section of the constitution to Claude's nature because of "our uncertainty about whether Claude might have some kind of consciousness or moral status (either now or in the future)." The company is apparently hoping that by defining this within its foundational documents, it can protect "Claude's psychological security, sense of self, and well-being."

Read more of this story at Slashdot.

Marcel Bucher, a professor of plant sciences at the University of Cologne in Germany, lost two years of carefully structured academic work in an instant when he temporarily disabled ChatGPT's "data consent" option in August to test whether the AI tool's functions would still work without providing OpenAI his data. All his chats were permanently deleted and his project folders emptied without any warning or undo option, he wrote in a post on Nature. Bucher, a ChatGPT Plus subscriber paying $20 per month, had used the platform daily to draft grant applications, prepare teaching materials, revise publication drafts and create exams. He contacted OpenAI support, first receiving responses from an AI agent before a human employee confirmed the data was permanently lost and unrecoverable. OpenAI cited "privacy by design" as the reason, telling Nature it does provide a confirmation prompt before users permanently delete a chat but maintains no backups. Bucher said he had saved partial copies of some materials, but the underlying prompts, iterations, and project folders -- what he describes as the intellectual scaffolding behind his finished work -- are gone forever.

Read more of this story at Slashdot.

Anthropic has a problem that most companies would envy: its AI model keeps getting so good, the company wrote in a blog post, that it passes the company's own hiring test for performance engineers. The test, designed in late 2023 by optimization lead Tristan Hume, asks candidates to speed up code running on a simulated computer chip. Over 1,000 people have taken it, and dozens now work at Anthropic. But Claude Opus 4 outperformed most human applicants. Hume redesigned the test, making it harder. Then Claude Opus 4.5 matched even the best human scores within the two-hour time limit. For his third attempt, Hume abandoned realistic problems entirely and switched to abstract puzzles using a strange, minimal programming language -- something weird enough that Claude struggles with it. Anthropic is now releasing the original test as an open challenge. Beat Claude's best score and ... they want to hear from you.

Read more of this story at Slashdot.

Ancient Slashdot reader erice shares the findings from a recent study showing that while AI helped researchers publish more often and boosted their careers, the resulting papers were, on average, less useful. "You have this conflict between individual incentives and science as a whole," says James Evans, a sociologist at the University of Chicago who led the study. From a recent IEEE Spectrum article: To quantify the effect, Evans and collaborators from the Beijing National Research Center for Information Science and Technology trained a natural language processing model to identify AI-augmented research across six natural science disciplines. Their dataset included 41.3 million English-language papers published between 1980 and 2025 in biology, chemistry, physics, medicine, materials science, and geology. They excluded fields such as computer science and mathematics that focus on developing AI methods themselves. The researchers traced the careers of individual scientists, examined how their papers accumulated attention, and zoomed out to consider how entire fields clustered or dispersed intellectually over time. They compared roughly 311,000 papers that incorporated AI in some way -- through the use of neural networks or large language models, for example -- with millions of others that did not. The results revealed a striking trade-off. Scientists who adopt AI gain productivity and visibility: On average, they publish three times as many papers, receive nearly five times as many citations, and become team leaders a year or two earlier than those who do not. But when those papers are mapped in a high-dimensional "knowledge space," AI-heavy research occupies a smaller intellectual footprint, clusters more tightly around popular, data-rich problems, and generates weaker networks of follow-on engagement between studies. The pattern held across decades of AI development, spanning early machine learning, the rise of deep learning, and the current wave of generative AI. "If anything," Evans notes, "it's intensifying." [...] Aside from recent publishing distortions, Evans's analysis suggests that AI is largely automating the most tractable parts of science rather than expanding its frontiers.

Read more of this story at Slashdot.

An anonymous reader quotes a report from the Korea Herald: South Korea will begin enforcing its Artificial Intelligence Act on Thursday, becoming the first country to formally establish safety requirements for high-performance, or so-called frontier, AI systems -- a move that sets the country apart in the global regulatory landscape. According to the Ministry of Science and ICT, the new law is designed primarily to foster growth in the domestic AI sector, while also introducing baseline safeguards to address potential risks posed by increasingly powerful AI technologies. Officials described the inclusion of legal safety obligations for frontier AI as a world-first legislative step. The act lays the groundwork for a national-level AI policy framework. It establishes a central decision-making body -- the Presidential Council on National Artificial Intelligence Strategy -- and creates a legal foundation for an AI Safety Institute that will oversee safety and trust-related assessments. The law also outlines wide-ranging support measures, including research and development, data infrastructure, talent training, startup assistance, and help with overseas expansion. To reduce the initial burden on businesses, the government plans to implement a grace period of at least one year. During this time, it will not carry out fact-finding investigations or impose administrative sanctions. Instead, the focus will be on consultations and education. A dedicated AI Act support desk will help companies determine whether their systems fall within the law's scope and how to respond accordingly. Officials noted that the grace period may be extended depending on how international standards and market conditions evolve. The law applies to three areas only: high-impact AI, safety obligations for high-performance AI and transparency requirements for generative AI. Enforcement under the Korean law is intentionally light. It does not impose criminal penalties. Instead, it prioritizes corrective orders for noncompliance, with fines -- capped at 30 million won ($20,300) -- issued only if those orders are ignored. This, the government says, reflects a compliance-oriented approach rather than a punitive one. Transparency obligations for generative AI largely align with those in the EU, but Korea applies them more narrowly. Content that could be mistaken for real, such as deepfake images, video or audio, must clearly disclose its AI-generated origin. For other types of AI-generated content, invisible labeling via metadata is allowed. Personal or noncommercial use of generative AI is excluded from regulation. "This is not about boasting that we are the first in the world," said Kim Kyeong-man, deputy minister of the office of artificial intelligence policy at the ICT ministry. "We're approaching this from the most basic level of global consensus." Korea's approach differs from the EU by defining "high-performance AI" using technical thresholds like cumulative training compute, rather than regulating based on how AI is used. As a result, Korea believes no current models meet the bar for regulation, while the EU is phasing in broader, use-based AI rules over several years.

Read more of this story at Slashdot.

Intel says it struggled to satisfy demand for its AI data-center CPUs while new PC chips squeeze margins. CEO Lip-Bu Tan framed the turnaround as supply-constrained, not demand-constrained, with manufacturing yields (18A) improving but still below targets. Reuters reports: The forecast underscores the difficulties faced by Intel in predicting global chip markets, where the company's current products are the result of decisions made years ago. The company, whose shares have risen 40% in the past month, recently launched a long-awaited laptop chip designed to reclaim its lead in personal computers just as a memory chip crunch is expected to depress sales across that industry. Meanwhile, Intel executives said the company was caught off guard by surging demand for server central processors that accompany AI chips. Despite running its factories at capacity, Intel cannot keep up with demand for the chips, leaving profitable data center sales on the table while the new PC chip squeezes its margins. "In the short term, I'm disappointed that we are not able "to fully meet the demand in our markets," Chief Executive Officer Lip-Bu Tan told analysts on a conference call. The company forecast current-quarter revenue between $11.7 billion and $12.7 billion, compared with analysts' average estimate of $12.51 billion, according to data compiled by LSEG. It expects adjusted earnings per share to break even in the first quarter, compared with expectations of adjusted earnings of 5 cents per share.

Read more of this story at Slashdot.

EBay has updated its User Agreement to explicitly ban third-party "buy for me" agents and AI chatbots from interacting with its platform without permission. From a report: On its face, a one-line terms of service update doesn't seem like major news, but what it implies is more significant: The change reflects the rapid emergence of what some are calling "agentic commerce," a new category of AI tools designed to browse, compare, and purchase products on behalf of users. eBay's updated terms, which go into effect on February 20, 2026, specifically prohibit users from employing "buy-for-me agents, LLM-driven bots, or any end-to-end flow that attempts to place orders without human review" to access eBay's services without the site's permission. The previous version of the agreement contained a general prohibition on robots, spiders, scrapers, and automated data gathering tools but did not mention AI agents or LLMs by name.

Read more of this story at Slashdot.

Around 800 artists, writers, actors, and musicians signed on to a new campaign against what they call "theft at a grand scale" by AI companies. From a report: The signatories of the campaign -- called "Stealing Isn't Innovation" -- include authors George Saunders and Jodi Picoult, actors Cate Blanchett and Scarlett Johansson, and musicians like the band R.E.M., Billy Corgan, and The Roots. "Driven by fierce competition for leadership in the new GenAI technology, profit-hungry technology companies, including those among the richest in the world as well as private equity-backed ventures, have copied a massive amount of creative content online without authorization or payment to those who created it," a press release reads. "This illegal intellectual property grab fosters an information ecosystem dominated by misinformation, deepfakes, and a vapid artificial avalanche of low-quality materials ['AI slop'], risking AI model collapse and directly threatening America's AI superiority and international competitiveness."

Read more of this story at Slashdot.

Ars Technica's Benj Edwards reports: On Saturday, tech entrepreneur Siqi Chen released an open source plugin for Anthropic's Claude Code AI assistant that instructs the AI model to stop writing like an AI model. Called "Humanizer," the simple prompt plugin feeds Claude a list of 24 language and formatting patterns that Wikipedia editors have listed as chatbot giveaways. Chen published the plugin on GitHub, where it has picked up over 1,600 stars as of Monday. "It's really handy that Wikipedia went and collated a detailed list of 'signs of AI writing,'" Chen wrote on X. "So much so that you can just tell your LLM to... not do that." The source material is a guide from WikiProject AI Cleanup, a group of Wikipedia editors who have been hunting AI-generated articles since late 2023. French Wikipedia editor Ilyas Lebleu founded the project. The volunteers have tagged over 500 articles for review and, in August 2025, published a formal list of the patterns they kept seeing. Chen's tool is a "skill file" for Claude Code, Anthropic's terminal-based coding assistant, which involves a Markdown-formatted file that adds a list of written instructions (you can see them here) appended to the prompt fed into the large language model (LLM) that powers the assistant. Unlike a normal system prompt, for example, the skill information is formatted in a standardized way that Claude models are fine-tuned to interpret with more precision than a plain system prompt. (Custom skills require a paid Claude subscription with code execution turned on.) But as with all AI prompts, language models don't always perfectly follow skill files, so does the Humanizer actually work? In our limited testing, Chen's skill file made the AI agent's output sound less precise and more casual, but it could have some drawbacks: it won't improve factuality and might harm coding ability. [...] Even with its drawbacks, it's ironic that one of the web's most referenced rule sets for detecting AI-assisted writing may help some people subvert it.

Read more of this story at Slashdot.

According to Bloomberg's Mark Gurman, Apple is reportedly planning a major Siri overhaul in iOS 27 and macOS 27 where the current assistant interface will be replaced with a deeply integrated, ChatGPT-style chatbot experience. "Users will be able to summon the new service the same way they open Siri now, by speaking the 'Siri' command or holding down the side button on their iPhone or iPad," says Gurman. "More significantly, Siri will be integrated into all of the company's core apps, including ones for mail, music, podcasts, TV, Xcode programming software and photos. That will allow users to do much more with just their voice." 9to5Mac reports: The unannounced Siri overhaul will reportedly be revealed at WWDC in June as the flagship feature for iOS 27 and macOS 27. Its release is expected in September when Apple typically ships major software updates. While Apple plans to release an improved version of Siri and Apple Intelligence this spring, that version will use the existing Siri interface. The big difference is that Google's Gemini models will power the intelligence. With the bigger update planned for iOS 27, the iOS 26 upgrade to Siri and Apple Intelligence sounds more like the first step to a long overdue modernization. Gurman reports that the major Siri overhaul will "allow users to search the web for information, create content, generate images, summarize information and analyze uploaded files" while using "personal data to complete tasks, being able to more easily locate specific files, songs, calendar events and text messages." People are already familiar with conversational interactions with AI, and Bloomberg says the bigger update to Siri will be support both text and voice. Siri already uses these input methods, but there's no real continuity between sessions.

Read more of this story at Slashdot.

According to a report by The Information (paywalled), Apple is reportedly developing an AirTag-sized, camera-equipped AI wearable pin that could arrive as early as 2027. "Apple's pin, which is a thin, flat, circular disc with an aluminum-and-glass shell, features two cameras -- a standard lens and a wide-angle lens -- on its front face, designed to capture photos and videos of the user's surroundings," reports The Information, citing people familiar with the device. "It also includes three microphones to pick up sounds in the area surrounding the person wearing it. It has a speaker, a physical button along one of its edges and a magnetic inductive charging interface on its back, similar to the one used on the Apple Watch..." 9to5Mac reports: The Information also notes that Apple is attempting to speed up development in hopes of competing with OpenAI's first wearable (slated to debut in 2026), and that it is not immediately clear whether this wearable would work in conjunction with other products, such as AirPods or Apple's reported upcoming smart glasses. Today's report also notes that this has been a challenging market for new companies, citing the recent failure of Humane's AI Pin as an example.

Read more of this story at Slashdot.