OpenAI’s latest response defends its firing of three safety researchers as justified by a “significant breach of trust.” [S2][S4]
The company explicitly says the dismissals were not about raising safety concerns or speaking out. [S1][S2][S3][S4][S5]
The three former researchers say instead that they were fired for prioritizing or raising AI-safety concerns. [S1][S3][S4][S5]
They also warn that the dismissals could chill safety-related speech and collaboration inside OpenAI. [S1][S5]
On October 2, OpenAI told the BBC that its investigation had confirmed mishandling of sensitive information outside established procedures. [S3]
In its Friday response, OpenAI said a thorough investigation found violations of clear policies governing sensitive information. [S2]
OpenAI said the breaches went beyond the researchers’ published account but did not provide details. [S4]
Verification result: verified, because five independent reporting families directly confirm the narrow central claim that OpenAI disputes the researchers’ account, with no validated contradiction to that claim. [S1][S2][S3][S4][S5]
The available reporting cannot resolve the actual motive for the firings because OpenAI did not detail the alleged breaches or answer specific questions about the policies and circumstances. [S4][S5]
