LeCun: Proving Theorems Doesn't Mean AI Is Beyond Human Level at Math
ylecun · x · 2026-09-16
Responding to claims that AI has surpassed humans at math, Yann LeCun argues that being better at searching for and writing down formal theorem proofs is just one automatable "mechanical task," not the whole of mathematics.
He compares it to arithmetic or computing integrals: the core of mathematical work is inventing new concepts, frameworks, abstractions, and definitions, and formulating conjectures — which requires intuition and creativity that current AI systems do not yet have.
More from AGI Musings
- Long-read: AI's real value comes after the efficiency era—infra glut, agent tooling boom — yangyi · 2026-11-02
- Security's attack-defense automation gap mirrors math: proofs are easy, conjectures are hard — mboehme_ · 2026-11-02
- vikhyat predicts alternating compute oversupply and undersupply on the road to AGI — TheZachMueller · 2026-11-02
- Sam Altman: AI has surpassed top scientists, and 2026 will be a turning point year — kimmonismus · 2026-11-02
- DeepSeek design doc: a person can only follow 3-5 agent sessions at once, attention is the real bottleneck — heyshrutimishra · 2026-11-02
- AI researcher: we have no reliable way to tell whether an AI system is conscious — dioscuri · 2026-11-02