OpenAI safety leader quits, calls company culture broken

A senior safety leader at OpenAI has resigned this week with a stark public warning that the company's internal culture is fundamentally broken, adding to a string of high-profile departures from the safety side of the organisation. The resignation lands as Anthropic published detailed reporting on its own efforts to embed moral reasoning directly into its Claude models, sharpening the public contrast between the two leading AI labs on safety governance. The timing matters: both companies are racing to ship increasingly autonomous systems, and who holds the safety wheel inside each lab is no longer an abstract question.