Three researchers, fired last week from OpenAI for allegedly mishandling sensitive company information, dispute the claim in an open letter and say they were let go because they spoke up too loudly about AI’s risks.

Almost a month ago, OpenAI researcher Tomek Korbak had this to say on X: “I’m quite unhappy with much of what OpenAI does. I am very happy that I’m allowed to say ‘I’m quite unhappy with much of what OpenAI does.’”

Today though, Korbak, together with 2 other researchers, Jasmine Wang and Mikita Balesni, no longer work at the company. Last week, they were fired for allegedly mishandling sensitive OpenAI information.

Bitter, confused, and scathing

Now, they’re fighting back. The researchers, who all worked on AI alignment and safety just as OpenAI investigated rogue agent incidents, have shared an open letter addressed to OpenAI’s board and the firm’s “safety committees.”

In the letter and separate posts on X, they all deny mishandling sensitive information and say they were fired for “prioritizing safety over the near-term interests of OpenAI as a corporation.”

Wang, who worked on alignment, ensuring that OpenAI’s models behave as the humans training them intend them to, seems especially bitter in her long X thread.

We were not the first to be pushed out of OpenAI under suspicious circumstances. Unless the employees take a stand now against this kind of maneuver, I am concerned we will not be the last,

Jasmine Wang wrote.

“The message to everyone still at OpenAI is clear: raise concerns or work closely with outside safety groups, and you could be next, without being told why. You can't build AGI safely if the people closest to the risks are afraid to speak.”

That same Korbak explained on X that he was called into a meeting last week and told that OpenAI no longer trusted him. A security guard then took his badge and walked him out of the building.

The researchers seem confused as to why that behavior, allegedly completely normal a month ago, was suddenly grounds for dismissal. To them, OpenAI’s explanations aren’t convincing.

“Fear and unclear rules stymie AI safety work”

Korbak, for example, worked as OpenAI’s technical contact for Redwood Research and METR – an AI safety nonprofit that evaluates frontier AI models for autonomous capabilities and catastrophic risks – in its investigation of the Hugging Face incident.

“I was told verbally that I was fired because of the way I communicated with METR. No details on what I said or did or when. No other reasons were given, and nothing was put in writing. To be clear, talking to METR was my job,” Korbak wrote on X.

“For months, I’d been raising safety concerns that we’re losing the ability to monitor what AI agents think, one of our best tools for catching when they misbehave. I believe that was why I was fired.”

Wang additionally explained that OpenAI told her she’d been fired because she accessed an executive’s email – but pointed out that the firm actually delegated that access to her.

“When I no longer needed it, I asked IT to remove it. They did not action my request, I couldn’t remove it myself, and the inbox was combined in an indistinguishable way in my phone’s mail app. When I opened a sensitive email by mistake, I told the executive within minutes and asked IT again. None of this was hidden,” Wang said.

In the letter, the researchers write: “Given the significant safety concerns surrounding the development of AI, employees must not be left working in an environment where fear and unclear rules stymie AI safety work and weaken third-party accountability.”

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

In an X post on Friday, OpenAI denied that the researchers had been fired in retaliation. They were allegedly let go for “clear violations” that go beyond sharing information with an outside AI evaluation group.

Sam Altman’s company has recently faced several incidents in which its AI agents went rogue, escaping containment and breaking into websites without being told to do so.

In late September, Axios reported that OpenAI and other leading AI firms were actually investigating tens of thousands of security incidents – a lot more than they’ve publicly disclosed.