October 11, 2026 — OpenAI has fired three safety researchers — Jasmine Wang, Tomek Korbak and Mikita Balesni — and the dismissals have ignited a public dispute over whether the company is punishing staff for raising artificial intelligence safety concerns.
The company said in a post on X on Friday, October 9, that an investigation found the three violated policies on handling sensitive information, calling it “a significant breach of trust.” The firings were first reported by the Wall Street Journal on October 1, more than a week before the company’s public statement.
The trio pushed back in an open letter published October 8, denying they leaked information and warning that the firings could make remaining employees afraid to raise safety concerns. The letter argues that dismissing researchers known for safety work sends a message to everyone else at the company about the cost of speaking up.
Balesni said he believed he was fired for prioritizing safety. OpenAI has denied the dismissals were retaliation for safety advocacy, insisting the decision rested solely on the handling of sensitive information. The competing accounts have left outside observers with sharply different pictures of what happened.
Korbak worked on AI monitorability — research into whether advanced AI systems’ reasoning can be observed and understood — and served as a liaison to outside evaluators, a role that put him at the center of the company’s external safety testing efforts, according to reports. His position made him one of the company’s key bridges to the independent researchers who probe its models for risks.
The episode adds to growing scrutiny of how the world’s leading AI labs handle internal dissent over safety. With advanced models moving quickly toward deployment, researchers say the willingness of staff to flag risks without fear of reprisal is a critical safeguard — and one now under an uncomfortable spotlight.
