Fired OpenAI safety researchers deny misconduct allegations and warn of a chilling effect on culture.

Three recently fired OpenAI safety researchers have published an open letter denying claims that they mishandled sensitive company information. Jasmine Wang, Tomek Korbak, and Mikita Balesni were dismissed last week after the company alleged they shared confidential details with an outside safety organization. In their letter to OpenAI governance committees, the researchers disputed the claims and denied involvement in a recent media leak.

Wang also explained on social media that her access to an executive email inbox, cited by the company, was an unremoved administrative artifact she reported immediately upon opening by mistake. The researchers warned that their abrupt dismissal creates a chilling effect on internal culture, leaving remaining employees afraid to collaborate with outside experts or raise safety concerns.

They argued that open dialogue with external evaluators is essential for addressing risks. OpenAI has not formally responded to the letter, but an internal memo shared with reporters stated that the researchers were not fired for raising safety concerns or speaking out.