Fired OpenAI safety researchers dispute misconduct claims, warn of chilling effect
Three researchers OpenAI dismissed say an open letter denies the company’s account of how they handled research information.
TL;DR
- TechCrunch names Jasmine Wang, Tomek Korbak and Mikita Balesni as the three safety researchers OpenAI fired. [1]
- The three published an open letter denying that they mishandled sensitive information, and warned of a chilling effect on safety work. [1]
- OpenAI has described a pattern of misconduct that goes beyond sharing information with an outside evaluation group. The dispute is unresolved in public. [1]
TechCrunch reports that Jasmine Wang, Tomek Korbak and Mikita Balesni, three safety researchers OpenAI fired last week, have published an open letter. The letter denies the company’s claim that they mishandled sensitive information outside established channels. [1] [1]
The researchers warn that the dismissals are creating a chilling effect on OpenAI’s AI safety culture. TechCrunch presents that warning as theirs, not as a finding by the newsroom. [1] [1]
OpenAI, in the account TechCrunch cites, described a pattern of misconduct and a clear violation of its policies on mishandling research information, and said the issue goes beyond sharing information with an outside AI evaluation group. OpenAI did not directly answer TechCrunch’s questions on that point in the report. [1] [1]
The story was heavily discussed on Hacker News. That discussion does not adjudicate the letter. What is confirmed is that the firing, the letter and the company’s misconduct framing are now on the public record. What is open is which account of the information-handling is accurate. [1,2] [1] [2]
Why it matters
A public break between a lab and its safety researchers is evidence about how that lab handles internal dissent, not only about one personnel file.
Editor's note
Names and the contents of the dispute are taken from TechCrunch’s account of the letter and of OpenAI’s statement. This brief does not choose a side.