Ex-OpenAI/Anthropic pretraining researcher quits, says labs are 'gambling our lives' on superintelligence
inductionheads · x · 2026-09-10
Hilbert Spaess, who spent three years doing pretraining research at both OpenAI and Anthropic, announced his resignation from Anthropic, accusing both labs of acting irresponsibly and "racing straight to self-improving superintelligence and gambling with our lives." The post resurfaced a $1,000 Long Bets wager between skeptic Michael Shermer and Harvard astronomer Avi Loeb over whether confirmed alien visitation will be disclosed by 2030. The core item is the safety-critical departure and its accusations.
More from AGI Musings
- AI Researcher Pushes Back on RSI Hype: Recent Events Provide ~0 Evidence on the Cruxes — NathanpmYoung · 2026-09-10
- Neil Chilson: doomer AI arguments fail on complex systems in three ways — neil_chilson · 2026-09-10
- Doomer argument rests on a sleight-of-hand: conflating intelligence with world-steering — neil_chilson · 2026-09-10
- The doomers' core flaw: not understanding complex systems — neil_chilson · 2026-09-10
- Neil Chilson lays out the doomer case: unalignable AI will crush humanity — neil_chilson · 2026-09-10
- Neil Chilson: doomer argument hinges on redefining intelligence as control — neil_chilson · 2026-09-10