Three safety researchers fired abruptly

Mikita Balesni, Tomek Korbak and Jasmine Wang, who worked on AI safety and alignment at OpenAI, were fired last week and released a letter Thursday explaining their terminations.

They say they were fired for talking to safety groups

Balesni said he was told the company no longer trusts him because he was speaking too much to third party safety organizations.

“To be clear, talking to METR was my job”

OpenAI says they violated information policies

OpenAI's spokesperson said an internal investigation found the three employees violated company policies regarding access to and handling of sensitive information. The employees received only verbal explanations for their firings.

Researchers warn of chilling effect on safety culture

In their letter titled OpenAI cannot make AI safe on its own, the three wrote that colleagues would now be afraid to speak and operate in ways that were integral to working at OpenAI until…

“Terminations such as ours, executed and communicated so abruptly, are chilling the open culture OpenAI has prized in the past”

OpenAI leader backs their safety recommendations

The researchers recommended embedding third-party safety auditors within OpenAI, preserving model monitorability, and supporting open dialogue between safety researchers and the broader safety ecosystem.