Ex-OpenAI staff say they were fired for raising safety concerns

What the report says

Three former OpenAI researchers, Mikita Balesni, Tomek Korbak and Jasmine Wang, said they were dismissed after raising concerns about AI safety and speaking with outside experts. In an open letter and social media posts reported by Al Jazeera on 9 Oct. 2026, they argued the firings could create a “chilling” effect inside the San Francisco-based company and make colleagues less willing to question practices or report risks.

OpenAI rejected that explanation, saying the employees were let go over misconduct involving company information and a broader breach of trust, not because they raised safety issues. The company said it still encourages candid debate on research and safety, and that it valued the trio’s contributions. The former staff members denied breaking policy and said external safety engagement fit their roles.

The dispute lands amid wider industry debate over how to manage advanced AI systems and their potential risks. Al Jazeera noted that OpenAI and rivals such as Anthropic have called for coordinated safety measures and stronger safeguards, while governments and advocates continue to disagree over how much oversight is needed. The case underscores the tension between corporate control, internal dissent and public accountability in a fast-moving field where safety concerns are increasingly central.