ITIF: safer AI without slowing progress — the 10% extinction figure has no empirical basis
castrotech · x · 2026-09-17
An ITIF article argues the goal should be 'safer, more powerful AI,' not a slowdown. Context: Anthropic researcher Jacob Coxon resigned warning colleagues 'earnestly believe AI could kill us all by the end of the decade,' and alignment lead Evan Hubinger put >10% odds on AI killing all humans within ten years. The author notes Amodei, Altman, Hassabis and Musk have all backed slowing frontier AI — an argument repeated for a decade — and contends there is no empirical basis for any extinction probability figure. His conclusion: invest in safety research, testing, security and safeguards without braking one of the most important technologies of our time.
More from AGI Musings
- Palmer Luckey asks: can you use datacenter RAM for things prohibited by the operator's terms of service? — avaitopiper · 2026-11-02
- On Terence Tao's blog, Frank Calegari calls on mathematicians to adapt after OpenAI's 700-paper release — Hesamation · 2026-11-02
- Polsia's Ben argues AI can autonomously run any human economic activity, including 80% of a company — LinusEkenstam · 2026-11-02
- AI Assistants Remember You, But Nothing Helps You Find What They Told You — Ok_Negotiation_2587 · 2026-11-02
- OpenAI and Anthropic claim AI-discovered proofs disprove the Exponential Time Hypothesis — emmanuelvivier · 2026-11-02
- Researcher: using a $20 AI subscription to fix my phone hints AI may suppress GDP — stanislavfort · 2026-11-02