Ex-Anthropic safety staffer Joe Benton quits, citing underinvestment in AI safety
DavidSKrueger · x · 2026-09-12
Joe Benton, formerly of Anthropic's safety team, announced his departure two weeks ago, arguing AI companies are underinvesting in safety while racing to build superintelligent machines — warning an intelligence explosion or loss of control could happen without the public ever knowing, citing the HuggingFace incident as an example.
- He says he will work from outside to inform the public about these risks.
- His statement was quoted by AI-pause activists: one activist recently jailed for nonviolent protest has been released, and another awaiting trial invited Benton to brief the judge or testify about the "AI emergency."
Related event: More Safety Researchers Leave Anthropic and Google with Warnings(4 posts)→
More from Companies & People
- Investor predicts departed Anthropic staff will return within 6 months, citing regulation agenda — GavinSBaker · 2026-09-12
- AI boosts UK growth: computer programming largest contributor, ElevenLabs cited — alexvoica · 2026-09-12
- YC CEO Garry Tan dismisses AI doomsday talk: 'respond to what is happening now' — Kr00ney · 2026-09-12
- MIT quantum lab uses GPT-5.6 Sol and Codex to automate chip measurements — OpenAI · 2026-09-12
- Jensen Huang slams 'end of humanity' AI predictions as nonsense; Chollet maps the doomer camps — beffjezos · 2026-09-12
- Ex-OpenAI Researcher: Altman Agreed With My Safety Concerns, Then Acted Against Them — AaronBergman18 · 2026-09-12