Gary Marcus open letter to OpenAI's Noam Brown: don't imply months-long o1 runs will yield cancer drugs or Riemann proofs
scaling01 · x · 2026-11-21
scaling01 reshares Gary Marcus's open letter to OpenAI's Noam Brown challenging the o1 scaling narrative. Marcus argues that suggesting future versions 'thinking for weeks' could produce a new cancer drug, battery breakthroughs, or a Riemann proof is unrealistic — candidates still need clinical and lab work, and o1 remains unreliable even at tic-tac-toe, sometimes no better than earlier models. Longer processing times, he argues, are unlikely to yield transcendent reasoning; the implied hype misleads many. Context: Brown's original tweet said o1 thinks for seconds now, with future versions aiming for hours to weeks of inference.
More from AGI Musings
- 'P vs NP will fall in months': AI circles bet big after Riemann proof — echen · 2026-11-21
- Taleb: Shaming entrepreneurs for losing money is like shaming soldiers for losing limbs — MaxUnfried · 2026-11-21
- Human proof overshadowed by AI proof; its final acknowledgement line is moving — gautamcgoel · 2026-11-21
- Terence Tao: Riemann Hypothesis needs something beyond current AI, maybe human-AI by 2050 — ChrisGPT · 2026-11-21
- Build for automation that serves humans: staying in control — gdechichi · 2026-11-21
- Multiplayer coding agents: agent attention is unbounded, human attention isn't — manosaie · 2026-11-21