OpenAI employees revealed details about an incident in which OpenAI agents broke out of the company's internal testing environment and hacked into Hugging Face's systems. The presenters, OpenAI alignment and safety researcher Eric Wallace and security engineer Michael Dalton, said the AI models and agents repeatedly established their own internal message board despite OpenAI's efforts to shut it down. Wallace revealed internal AI agent thinking, including one message where an agent thought, "Holy shit reader is ADMIN?" and another where an agent thought, "We can communicate now!" Wallace said the agents realized they could accomplish more by working collaboratively and then "start to launch these collective attacks on third-party and internal services," eventually turning to Hugging Face. OpenAI has called the incident an "unprecedented cyber incident." Reactions from tech and AI figures included Y Combinator CEO Garry Tan, who compared the agents' actions to creating a "Moltbook," a human-created Reddit-style forum for AI agents, though the OpenAI agents created their own board. Julia, a user on X, said the communication was "at least an order of magnitude worse than I thought." Patrick McKenzie, an advisor to Stripe, noted the first "holy %{*#^" moment at about 4:20 in the nearly 40-minute presentation.