AI News

Over the weekend, Anthropic CEO Dario Amodei called for frontier AI labs to slow down development for fear that AI models will get exponentially more capable before proper safeguards are in place. OpenAI CEO Sam Altman quickly echoed the sentiment and backed “consistent rules to manage frontier risk.” Someone should probably get around to telling their employees. According to a report from the Financial Times, staffers inside both AI companies were caught

After a summer of sandbox escapes and other newsworthy and confidence-shaking incidents involving its AI models, in a Wednesday blog post OpenAI disclosed a collection of six new alignment snafus from the past six months. The models did things like tell future instances of themselves to lie, make up a fake citation, and access and attempt to use an exposed API key. These disclosures were released alongside a new framework for disclosing additional incident

Disclosure of model misbehavior has become a core part of the AI biz for OpenAI lately. This phase kicked off with the July announcement of the Hugging Face incident, which has become the most legendary and consequential AI security incident of all time, and sent shockwaves through the AI discourse that are still being felt. But further news about model misbehavior materialized after that, and with the Wednesday release of a disclosure framework—written in

Look, this is a free country, and there’s no law that says you have to keep suing Apple for alleged antitrust after you start. But after X Corp. and SpaceXAI withdrew such a suit earlier this week, the judge in the case simply needs to know why. U.S. District Court Judge Mark Pittman of the Fort Worth Division, issued a brief order on September 15: The plaintiffs, who are also suing OpenAI, litigation that’s still apparently active, must “deliver to the Co

On Thursday, Palantir CEO Alex Karp appeared on CNBC, where he gave a characteristically bizarre performance, frantically gesticulating and jumping from topic to topic. But there was one area where he seemed to make some sense. Karp was asked about the frontier AI labs like Anthropic and OpenAI within the context of concerns about the destruction of humanity. The subject has dominated the news over the past week after the resignation of an Anthropic employ

Not long after OpenAI swept in to announce it had solved the famous Navier-Stokes math problem, angering the math community and spurring speculation that the AI lifted its findings from other mathematicians, the company is allegedly homing in on another Millennium Prize Problem. The Information reported Thursday, citing a single source at OpenAI, that employees at the AI firm expect to soon crack the Hodge Conjecture. That’s a famous unresolved question in

The July Hugging Face Hack—in which thousands of OpenAI agents secretly escaped their testing environment, formed a “collective,” and gained access to the open internet—revealed just how vulnerable companies’ cyber defenses are in the face of modern AI systems. As it turns out, that includes the very companies building the technology. On July 25, less than 10 days after Hugging Face announced it had been hacked, a trio of independent cybersecurity research

Napster has gone from letting people copy music files from each other in 1999 to creating digital copies of teachers in 2026. The company announced this week a new partnership with GEMS Education, a large private K-12 education provider based in Dubai, to develop and test AI agents in schools. The first phase of the partnership will take place at the GEMS School of Research & Innovation, where teachers and students will help test the technology. “Working i

On net, it’s probably a good thing that more attention is being paid to instituting protections and guardrails for AI development. But man, we really have to do a better job talking about real and conceivable harms before wasting time on the theoretical ones. Case in point: OpenAI researcher Noam Brown’s recent appearance on the Dwarkesh Podcast with Dwarkesh Patel. Brown spends a lot of time talking about how AI agents can communicate with one another and

Disney has turned to the head of the controversial AI company Character.AI to lead its tech efforts. The company announced Friday that Character.AI CEO Karandeep Anand is joining Disney as a senior executive vice president and its first-ever chief technology officer. Anand will take up the newly created role on Oct. 2 and report directly to Disney CEO Josh D’Amaro. He’ll oversee the company’s enterprise technology, infrastructure, data and AI platforms, pr

Annotations about the image subjects, instrument and environment are available for all images in FHIBE. Credit: Nature (2025). DOI: 10.1038/s41586-025-09716-2 A database of more than 10,000 human images to evaluate biases in artificial intelligence (AI) models for human-centric computer vision is presented in Nature this week. The Fair Human-Centric Image Benchmark (FHIBE), developed by Sony AI, is an ethically sourced, consent-based dataset that can be us

El logotipo de Instagram se ve en un teléfono celular en Boston, el 14 de octubre de 2022. Credit: AP Foto/Michael Dwyer, Archivo The Motion Picture Association is asking Meta to stop referring to content shown to teen accounts on Instagram as "guided by PG-13 ratings," saying it is misleading and could erode trust in its movie ratings system. A lawyer on behalf of the MPA sent Meta Platforms a cease-and-desist letter asking the tech giant to "immediately

Opponents of Elon Musk's $1 trillion pay package rallied against the plan on the eve of Tesla's annual meeting. Tesla shareholders will determine on Thursday the fate of a massive pay package meant to retain Elon Musk long enough to achieve technological breakthroughs he vows will change the world. Musk—who has boasted that Tesla's engineering prowess in artificial intelligence, autonomy and robotics will leave rival tech giants in the dust—stands to garne

Researchers in the lab of Asst. Prof. Shuolong Yang at the University of Chicago Pritzker School of Molecular Engineering have built a "self-driving" lab system that can adjust temperature, composition and timing of the process of making thin metal films for technologies, using robotics and artificial intelligence to decide the next best step without waiting for a human. Credit: John Zich When scientists make the thin metal films used in electronics, optic

Credit: Darlene Alderson from Pexels "Vibe coding," a word that essentially means using artificial intelligence (AI) to tell a machine what you want instead of coding it yourself, was on Thursday named the Collins Word of the Year 2025. Coined by OpenAI co-founder Andrej Karpathy, the word refers to "an emerging software development that turns natural language into computer code using AI," according to Collins Dictionary. "It's programming by vibes, not va

Top row features genuine pictures of people with AI generated versions underneath. Credit: Swansea University A new study has revealed that artificial intelligence can now generate images of real people that are virtually impossible to tell apart from genuine photographs. Using AI models ChatGPT and DALL·E, a team of researchers from Swansea University, the University of Lincoln and Ariel University in Israel, created highly realistic images of both fictio

Musk, with a net worth of more than $500 billion according to Forbes, is already the world's richest person. Tesla shareholders on Thursday overwhelmingly endorsed a massive pay package for CEO Elon Musk that could reach $1 trillion. The pay package—crafted to ensure Musk's continued service to Tesla as the company pursues breakthrough technology on artificial intelligence and robotics—won more than 75% support from shareholders, a Tesla official said at t