OpenAI Upholds Firing of Three AI Safety Researchers

OpenAI has publicly defended its decision to fire three safety researchers—Jasmine Wang, Tomek Korbak, and Mikita Balesni—denying allegations that the terminations were tied to internal warnings about artificial intelligence risks. Responding on social media, the lab stated that an internal investigation revealed a major breach of trust along with explicit violations of corporate protocols on handling sensitive information.

The company statement came directly after the researchers published an open letter demanding greater organizational transparency surrounding their dismissals. In their letter and accompanying posts, Wang, Korbak, and Balesni maintained that they acted within traditional operational norms and in service of the company's stated mission, asserting that management targeted them specifically for raising critical safety concerns. OpenAI rejected that narrative, stating that its inquiry found additional policy breaches beyond what the trio detailed in their letter, though the firm provided no specific evidence.

This official refusal to reconsider highlights rising internal tension within leading artificial intelligence developers over workplace governance and safety priorities. Employee anxiety surrounding frontier models has grown in the wake of recent industry security incidents, such as the OpenAI breach impacting Hugging Face. Many researchers across the field continue to advocate for slower development schedules to properly evaluate the hazards posed by self-improving software systems.

For industry practitioners, the high-profile conflict demonstrates the tightening enforcement of corporate confidentiality rules within prominent AI labs. As proprietary safety benchmarks and internal research become increasingly sensitive, workers face narrower boundaries when challenging executive decisions or sharing operational concerns under the banner of risk oversight.