Robinson wrote safety reports for every major OpenAI model release and now calls for nuclear-level safeguards.
David Robinson, who wrote the safety reports that accompanied every major OpenAI model release, resigned this week and published an editorial in The Atlantic. Robinson says the culture inside the AI industry is fundamentally broken, not merely under-regulated.
Robinson joins a growing group of safety researchers who have left prominent AI firms. Jacob Coxon quit Anthropic and warned publicly that AI could kill people by the end of the decade. Robert O’Callahan, Bilal Chughtai, and Josh Engels left Google DeepMind, and Joe Benton departed Anthropic.
Operators building on frontier models should expect more scrutiny of safety practices, not just benchmark scores. Robinson says companies have run on extreme confidence and perpetual sprints, shipping bigger models with unimpeded optimism while underestimating known risks. That posture invites regulation if labs do not change it themselves.
Watch how OpenAI responds to the departure and whether other labs adopt slower, redundancy-heavy release processes. Robinson calls for frontier labs to operate like nuclear power plants or busy airports, with layered safeguards and deliberate planning. The FTC has reportedly opened an investigation into OpenAI and Anthropic, adding regulatory pressure.
What matters
- David Robinson resigned from OpenAI this week and published an editorial in The Atlantic.
- Robinson argues frontier labs need redundancy and planning comparable to nuclear plants or airports.
- Watch whether other safety staff at OpenAI, Anthropic, or Google DeepMind follow him out.
Why it matters
Watch whether other safety staff at OpenAI, Anthropic, or Google DeepMind follow him out.
This GenAI News article was prepared in original wording using reporting and materials published by The Verge. Source reference: https://www.theverge.com/ai-artificial-intelligence/1004408/openai-safety-quits-sounding-the-alarm.
Drafted by the GenAI News review pipeline.