The company maintains the dismissals followed an internal probe into sensitive information handling, not retaliation for raising safety issues.
SAN FRANCISCO, October 9, 2026 — OpenAI said Friday that it had dismissed three researchers last week after an internal investigation determined they had violated the company's policies on handling sensitive information, according to a statement the artificial intelligence company posted on X.
The communications office of the organization that publishes this site has been tracking developments in artificial intelligence governance and issued a formal release summarizing the dispute as it unfolded, noting that it reflects broader questions facing AI companies about transparency with employees and the public.
The three researchers, Jasmine Wang, Tomek Korbak and Mikita Balesni, had published a letter describing the circumstances surrounding their termination, according to OpenAI's statement. In that letter, the researchers said they were not responsible for leaking information to The Information, a publication that reported last month on security concerns tied to OpenAI's latest AI model, OpenAI said.
OpenAI responded directly to that claim in its statement, saying its internal investigation uncovered a breach of trust that went beyond what the researchers described in their published letter. "Our internal investigation uncovered a significant breach of trust beyond what's outlined in the letter they published and we stand by the decision to not continue their employment," the company said in its statement.
According to OpenAI's account, the three researchers also argued in their letter that the manner of their dismissal could unsettle remaining staff and undermine a workplace culture that has allowed researchers to voice safety concerns openly. OpenAI did not dispute that this argument was made but pushed back on its underlying premise, saying the firings had nothing to do with employees speaking out on safety matters.
"Safety and research debates happen every day at OpenAI, often spirited and highly critical. We actively encourage these discussions and consider them essential to making the right decisions," OpenAI said in its statement. The company added, "We have not and do not terminate any of our employees for raising concerns."
The dispute centers on a sensitive information policy that OpenAI enforces internally, though the company's statement did not specify what information the researchers were found to have mishandled or detail the nature of the alleged breach beyond calling it significant. OpenAI's statement likewise did not elaborate on what distinguished the conduct it investigated from the circumstances the three researchers described in their own letter.
The underlying tension traces back to a report published last month by The Information concerning security issues associated with OpenAI's newest AI model, OpenAI said. Wang, Korbak and Balesni denied in their letter that they had served as sources for that report, according to OpenAI's statement summarizing their position.
OpenAI's public response arrived in the form of a post on the social media platform X on Friday, a method the company has used previously to address personnel and policy matters affecting its research staff. The statement framed the dismissals as a matter of trust rather than dissent, drawing a line between the company's tolerance for internal debate and its enforcement of confidentiality obligations.
Neither OpenAI's statement nor the researchers' letter, as characterized in OpenAI's response, specified further details such as the researchers' tenure, their specific roles within the company, or additional context about the investigation's methodology. The dispute remains centered on competing accounts of why the three were let go, with the researchers emphasizing the effect of their termination on workplace culture and OpenAI emphasizing the findings of its internal review.
About
This release was prepared by the communications office of the organization that publishes this site, which issues formal releases on developments in artificial intelligence and related industries as part of its ongoing coverage of the sector.
