Yudkowsky's lottery analogy: we know superintelligence won't want to keep you alive
inductionheads · x · 2026-09-14
Eliezer Yudkowsky distills his doomer logic with a lottery analogy: just as you can be certain a lottery ticket will lose without knowing the winning number, uncertainty about what machine superintelligence wants is itself proof it won't want to keep you alive.
The reposter challenges this, framing it as the entire foundation of doomerism — capturing the ongoing dispute over the structure of Yudkowsky's AI safety arguments.
More from AGI Musings
- The definition of AI 'slop' keeps shifting: from errors to whether humans deserve credit — RexDouglass · 2026-09-14
- AI Writing Is Fine: Dev Pushes Back on the AI-Text Backlash — intellectronica · 2026-09-14
- OpenAI Exec: Scientific Intelligence's Endgame Is New Industrial Capacity Anywhere — shyamalanadkat · 2026-09-14
- RL Pioneer Szepesvari: A Single Human In Charge Of AI Poses Low Extinction Risk — CsabaSzepesvari · 2026-09-14
- Peter Diamandis: AGI and ASI are amplifiers that multiply whatever you point them at — PeterDiamandis · 2026-09-14
- 25 Fields medalists including Terence Tao sign open letter on AI in mathematics — haider1 · 2026-09-14