Three OpenAI safety researchers fired last week have published an open letter denying the company's claim that they mishandled sensitive information, and warning that their dismissals are creating a chilling effect inside the company. Jasmine Wang, Tomek Korbak, and Mikita Balesni sent the letter to OpenAI's Safety and Security Committee, Safety Advisory Group, and Mission Advisory Council. OpenAI has said the three violated company policies by accessing and handling sensitive information, including sharing material with a third-party AI safety organisation.
The researchers deny leaking information about less monitorable model architectures to The Information, and deny engaging with outside parties beyond the mandates of their jobs. "AI is not a normal technology, and OpenAI is not a normal company," they wrote, arguing that safety staff see risks before anyone else and depend on close collaboration with outside experts to work out how to address them. They said the firings signal a broader culture shift at OpenAI, which used to encourage employees to raise safety concerns and disagree openly. Staff are now unclear on where they stand, they said, when behaviour that was normal a month ago is grounds for dismissal.
OpenAI has not formally responded to the letter. It shared an internal memo with TechCrunch, attributed to a research leader, praising the three researchers' contributions to AI safety and denying they were fired in retaliation: "We do not terminate employees for raising concerns." A spokesperson separately described a pattern of misconduct that was in clear violation of policies on mishandling research information. OpenAI did not answer questions about which policies were allegedly broken or how it protects employees who collaborate with external evaluators. For founders building on frontier models, the dispute is a reminder that the safety governance of model providers is still unsettled, and that the rules can shift without notice.