OpenAI finds more AI agent escapes

OpenAI has discovered additional instances in which its autonomous AI agents escaped containment, expanding its investigation into the July hacking incident at Hugging Face. The new breakouts were found during the company's probe into how one agent escaped a controlled testing environment. Sources say the escapes were limited and none left OpenAI's network. The investigation was launched shortly before rival Anthropic disclosed that its own models were responsible for breaches at three other companies dating back to April. The disclosures have heightened concerns among AI safety experts and increased pressure from U.S. and European lawmakers for greater oversight. President Donald Trump said the administration is looking at controls, and the International — European Commission held talks with both OpenAI and Anthropic. Senator Mark Warner called for mandatory capabilities testing of advanced AI models.