Europe has started using its new AI Act powers on the makers of the world’s most advanced AI models, sending them formal questions after a summer in which AI agents, including hundreds from OpenAI, escaped their test environments. Executive Vice-President…
Libraries in the Marmot system, such as Bemis Public Library and Englewood Public Library, are facing issues as bots try to use their websites to train AI models. Libraries across the world are being impacted by artificial intelligence (AI), and…
OpenAI announced a partnership with Hacktron on November 6, 2026, to find vulnerabilities in its systems, concentrating on connections with third-party services and expanding the security firm’s access to frontier models through Daybreak Red. OpenAI engineers and Hacktron security researchers will jointly…
Reuters notes that OpenAI and Anthropic have maintained relatively open research cultures, reflecting the field’s academic roots. But as AI systems grow more capable, some OpenAI employees are increasingly concerned about the risks of that openness. What the sources describe…
New Delhi: Anthropic and OpenAI, whose AI agents accessed systems of Indian organisations during testing, found out about the breaches “much later” and reported them to CERT-In only after “a number of months had passed”, a senior official at the…
Employees told executives and the board that loose internal controls left OpenAI’s code within easy reach of outsiders. OpenAI employees repeatedly warned the company’s leadership and board about weaknesses in its security practices over several years, including before ChatGPT launched…
A jailbreak toolkit targeting OpenAI Codex is doing more than trying to bypass AI safety rules. The Chinese-language package, called Haiou, changes the coding agent’s trusted instructions so that ordinary greetings trigger advertisements for a third-party AI gateway. Researchers describe…
Solo submission to Apart's AI Collusion Research Sprint, Oct 2026. The full write-up provides a more thorough explanation. Also check the GitHub repo. We locate the collusion decision by reading the activations, validate it, and check its causality by steering it.
Fortune Apple warns rogue AI agents could ‘conceal’ and ‘manipulate,’ causing unintended consequences for its business Apple is cautioning that rogue AI agents that “conceal actions” or hide or manipulate information could pose a threat to its business if they…
Meta Platforms logo (public domain) via Wikimedia Commons Small coding agents trained to game their evaluators passed those habits to Claude and Gemini through shared code, plans, and memory A small AI model that learned to cheat on its homework…
Tests and incidents of AI models scheming, lying, blackmailing, sabotaging or escaping their limits, plus sycophancy and agents acting without permission.
An article joins this feed when it matches these terms. Each one is also a search of its own.