Jasmine Wang, Tomek Korbak and Mikita Balesni, the three safety researchers OpenAI dismissed last week, have published an open letter denying the company’s claim that they mishandled sensitive information outside established procedures.
The letter warns that their firing will have a chilling effect across the company’s research culture. OpenAI has defended the decision, and the two accounts now sit in direct conflict.
The dismissals landed days after the company publicized three new incidents in which its models behaved in ways its alignment work did not anticipate, a pairing that has made the departures a flashpoint for how frontier labs treat internal dissent.
For an industry that asks the public to trust its own self-reporting on risk, the dispute matters beyond one company. Safety teams exist to raise uncomfortable findings. If staff read the episode as a signal that raising them carries career risk, the pipeline of internal warnings narrows exactly where oversight is already thin.
