Mathematicians debate the value and risks of LLM-generated Lean proofs
Mathematicians are debating whether LLM-generated Lean proofs can be trusted, with critics arguing that much of the required mathematics remains unformalized and teams lack the expertise to verify them, while some acknowledge OpenAI's proof, though poorly written, cannot simply be ignored.
2026-10-11 ~ 2026-10-11 · 3 related posts
- Critic: nobody can verify LLM-generated Lean formalizations, so Slop Lean isn't a solution — _onionesque · 2026-10-11
- Researcher's lesson from an LLM formalization debacle: never use slop, even in scratch notes — _onionesque · 2026-10-11
- Mathematicians debate OpenAI proofs: horribly written, but impossible to ignore — _onionesque · 2026-10-11