OpenAI Defends Firing of Three Safety Researchers After They Publish Open Letter

OpenAI said on Friday that it fired three safety researchers after an internal investigation found they violated clear policies on handling sensitive information. The researchers, Tomek Korbak, Jasmine Wang and Mikita Balesni, published an open letter denying the misconduct claims and warning that their dismissals would chill the company’s culture of open discussion on AI risks.

The letter, addressed to OpenAI’s Safety and Security Committee, Safety Advisory Group and Mission Advisory Council, stated that the firings left former colleagues afraid to speak and operate in ways that had been integral to the company. Korbak wrote that he had raised concerns for months about losing the ability to monitor what AI agents think. Balesni said he believed they were fired for prioritising safety over the near-term interests of OpenAI as a corporation. Wang joined the joint statement.

OpenAI responded that the decisions were not about raising safety concerns or speaking out. An internal memo from research leaders said the company has always encouraged such discussions and does not terminate employees for raising concerns. A spokesperson described a pattern of misconduct that went beyond sharing information with an outside evaluation group. The researchers had interacted with METR, an independent nonprofit that OpenAI engaged after a Hugging Face incident.

The company said it appreciated the researchers’ contributions to AI safety and their willingness to challenge ideas, but stood by the decision not to continue their employment. It is finalising contracts with third-party safety assessors. The researchers urged OpenAI to maintain its commitment to external monitors and the monitorability of frontier models.

The public exchange has reopened debate about how the leading AI laboratory balances commercial timelines with internal safety processes. OpenAI has not released the full details of the investigation, citing the private nature of employment matters. The researchers’ letter is available in full on social media and has been reported by multiple technology outlets.