OpenAI recursive self-improvement researcher warns 70% extinction risk within 3 years, sparking slowdown debate
ccerrato147 · x · 2026-09-11
Polymarket quotes an OpenAI recursive self-improvement researcher warning of a 70% chance of "human extinction" within 3 years absent a coordinated slowdown across labs.
Dave Shapiro pushed back, noting the doomers' six-month pause demand years ago never happened and nothing went wrong, calling such warnings attention-seeking. The thread captures the latest accelerationist-vs-slowdown clash in AI safety discourse.
More from AGI Musings
- DeepMind researcher: all of mathematics will soon be formalizable, with agents hunting proofs — ChrSzegedy · 2026-10-28
- AI reply bots are degrading X discourse, a 'Gresham's Law' for social replies — generativist · 2026-10-28
- Be prepared: your life's research problem could be solved by a stranger's idle AI prompt on a commute — fkasummer · 2026-10-28
- A fine-tuned 'cat obsession' control fools every AI welfare criterion — MirrorEthic_Anchor · 2026-10-28
- OpenAI researcher: frontier AI ideas can't be patched for safety after the fact — willdepue · 2026-10-28
- Big Short's Eisman: the whole market is now 'one trade' on AI — and it 'better succeed' — SumitGup · 2026-10-28