Gary Marcus: Riemann proof relied on symbolic AI and humans, not LLMs alone
GaryMarcus · x · 2026-11-21
Gary Marcus pushes back on critics over the Riemann hypothesis proof: he predicted in 2024 that o1 alone wouldn't solve it, and that neurosymbolic approaches like AlphaProof and AlphaGeometry were the right direction.
The result, he argues, validated both claims — the proof came from a neurosymbolic system working with a brilliant human for two months, not an AI solving it on its own, and Lean (symbolic) was essential. He accuses tech bros of ignoring human involvement and turning science into a partisan team sport, urging the field to return to real science.
More from AGI Musings
- Chollet flashback: from 5% on ARC1 to saturating ARC1-3, scaling01 says it's just RL — scaling01 · 2026-11-21
- A written RH proof would cost ~$5,000 in output tokens before overhead, dev estimates — felpix_ · 2026-11-21
- Was RH really solved by Claude? Skeptic asks why no professor spent $200k to try — felpix_ · 2026-11-21
- Riemann Hypothesis reportedly solved, echoing the key scenario in "If Anyone Builds It, Everyone Dies" — jammastergirish · 2026-11-21
- Martin Casado: humans compressed the universe into language, LLMs just traverse it — yasamanbb · 2026-11-21
- Founder jokes simulation exists because designers had unused token credits to burn — PaulYacoubian · 2026-11-21