Debate over Paul Christiano's '90 math proofs' AI progress graph: where are the first 30?
allTheYud · x · 2026-10-07
A debate broke out over the progress graph Paul Christiano used in his post suggesting AI has solved 90 math proofs. allTheYud flags the key issue: on this nice smooth curve, where are the 30 proofs that came before the 90, and how much earlier should they have appeared?
The core concern is measurement methodology: without a preregistered fixed problem set, the curve's smooth takeoff may be inflated by selective counting.
Related event: OpenAI Math Results Split the Math World Between Awe and Existential Doubt(47 posts)→
More from AGI Musings
- Steven Pinker on AI Doom and Consciousness: Why He Thinks Doomerism Fails — sapinker · 2026-10-08
- 'The best time to be a researcher': AI is crushing the old scarcity-based game — aran_nayebi · 2026-10-08
- A mathematician explains why to be excited, not frightened, by the AI mathematics revolution — StefanoGogioso · 2026-10-08
- What Three-Body Problem's longtermism teaches about facing an ASI-level threat — JacquesThibs · 2026-10-08
- COLM'26: MathDuels self-play benchmark co-evolves difficulty across 19 frontier models — AI4Code · 2026-10-08
- eigenrobot debates EA forecasting: quantitative beats vague "I'm unsure" — eigenrobot · 2026-10-08