Fired OpenAI researchers say they were let go for 'prioritising safety'

Three former OpenAI researchers are pushing back after being sacked for what the company described as mishandling sensitive information. Their version is very different: they say the real issue was that they kept pressing safety concerns inside the company.

Mikita Balesni said on X that he believed the firings happened because they put safety ahead of OpenAI’s near-term corporate interests. He posted that message alongside an open letter to the company, written with fellow ex-employees Tomek Korbak and Jasmine Wang. All three say they had been working in safety or alignment roles until recently.

OpenAI has stuck to its explanation. On 2 October, a spokesperson told the BBC that an investigation had “confirmed that these individuals mishandled sensitive information outside established company procedures.” The company repeated that position on Friday morning in a note from research leaders, saying the decisions were “not about raising safety concerns or speaking out” and that the internal probe found “a significant breach of trust.”

The dispute lands in the middle of a wider argument over AI safety and what advanced systems could do if they go wrong. In their letter, the former staffers said the firings had made employees at the company afraid to speak up or work in the way they had before. Korbak said he had spent months warning that OpenAI was losing the ability to monitor what AI agents think, which he called one of the company’s best tools for catching bad behaviour.

OpenAI also said it was finalising contracts with third-party safety assessors and would share details in the coming weeks. Wang, meanwhile, warned that unless employees push back against this kind of move, she feared more people could be forced out under similar circumstances.