AI safety researcher: make systems provably safe before release, no slowdown needed
ghadfield · x · 2026-11-11
ghadfield voices anxiety about rapid AI progress — disruption, chaos, the internet going down, banks crashing — but argues we are not helpless.
The action needed right now doesn't require a long-term slowdown: make sure these systems are safe enough before they go out into the world, and build the independent capacity to verify that safety.
More from AGI Musings
- AI progress is now a one-year forecast — the 'no track record' excuse is gone — peterwildeford · 2026-11-11
- Hot take: Grokipedia already beats Wikipedia, which will become 'a platform for boomers' — SydSteyerhart · 2026-11-11
- Poll: 70% say Anthropic's best internal model plus $50M can dominate any human expert — JacquesThibs · 2026-11-11
- David Patterson: Within a Year Most Will Believe SI Can Do All Jobs — davidpattersonx · 2026-11-11
- AI experts may be better at predicting RSI timelines than enterprise adoption — AndyMasley · 2026-11-11
- TaskNet: AI-Led Worker Interviews Add 79 Tasks Missing from O*NET, Raise AI Exposure for Lawyers and Devs — stanfordnlp · 2026-11-11