OpenAI safety lead David Robinson quits, calling company culture 'broken'

Key points

• Robinson spent 3.5 years at OpenAI, overseeing safety reports for 12 frontier model launches before leaving.
• He drafted OpenAI's current Preparedness Framework, the company's internal rulebook for model safety.
• In an Atlantic essay, he said staff were too busy sprinting to consider or make major changes.
• His exit follows OpenAI firing three researchers, reportedly for passing sensitive information to an outside safety group.

Why it matters for marketers: Marketers citing OpenAI's safety credentials should expect closer public scrutiny, since a former safety report lead now publicly questions the company's internal culture.