p(doom) debate: do most civilizations fail alignment before controlling ASI?
allTheYud · x · 2026-09-09
- allTheYud pushes back on >99% p(doom): there should be variance across civilizations, and humanity may be slightly worse than the median in incompetence and discoordination — most planets probably don't produce a Hitler figure
- tszzl counters that assuming the median is better is an "ungrounded religious conviction," while agreeing that variance means a >0 number of species would solve alignment
- A representative AI-safety debate on doom probabilities, the great filter, and whether ASIs would be non-expansionist
More from AGI Musings
- Hot take: AI is just masking human burnout for longer, and that's not good — alienelf · 2026-09-09
- Reddit user questions OpenAI's superintelligent agent tests: are monitoring and security sufficient? — wabawanga · 2026-09-09
- Anthropic researcher quits over AI safety, cites 10% extinction risk — Nexusyak · 2026-09-09
- antirez rebuts Terence Tao: AI proofs add to math knowledge, not subtract from it — antirez · 2026-09-09
- Terry Tao: AI's closure of ML was bad; in pure math it'd be a civilizational tragedy — KordingLab · 2026-09-09
- French Math Society president 'devastated' as AI solves a problem colleagues pondered for years — IgorCarron · 2026-09-09