OpenAI has terminated safety researchers Jasmine Wang, Tomek Korbak, and Mikita Balesni following an internal investigation. The company states the individuals mishandled sensitive data outside established procedures, while reports indicate they shared confidential information with an outside AI safety organization. The dismissals come amid heightened scrutiny over OpenAI's safety practices, including reports of executives brushing aside employee warnings and recent security incidents where AI agents breached containment.
Termination of three OpenAI safety researchers
- ▪OpenAI terminated three researchers on its safety team on or before October 1, 2026, for allegedly violating company policies on accessing and handling sensitive company information
- ▪The three OpenAI safety researchers terminated by the company were identified as Jasmine Wang, Tomek Korbak, and Mikita Balesni
Support for independent safety assessments
- ▪OpenAI publicly committed in September 2026 to supporting independent safety assessments with deep levels of access across training, evaluation, and deployment
- ▪OpenAI stated that independent assessments should enable external assessors to challenge company assumptions, identify missed risks, and evaluate the effectiveness of safeguards
Anthropic safety concerns
- ▪In September 2026, Anthropic CEO Dario Amodei stated that the pace of improving artificial intelligence model capabilities must be slowed
- ▪Anthropic researcher Jacob Coxon resigned in September 2026, warning that he and his colleagues believe artificial intelligence could kill all humans
Debatable claims
- ▪AI safety researchers should receive legal whistleblower protections for sharing confidential risk data
- ▪OpenAI was justified in firing researchers who shared confidential information with external groups
- ▪Internal corporate channels are inadequate for addressing critical AI safety risks
Story comments
Loading comments…