ArXiv paper: Lean verification of AI autoformalisation doesn't guarantee correct natural language proofs
RexDouglass · x · 2026-10-08
An arXiv paper (2610.08144, Bastounis, Circelli, Hansen) targets the 'AI autoformalisation + Lean verification' pipeline underpinning OpenAI's announced Navier-Stokes blow-up proof. Key points: resolving ambiguities in mathematical natural-language text—necessary for semantically faithful translation—sits arbitrarily high in the Solvability Complexity Index hierarchy (SCI = ∞), informally harder than any computational problem including Halting (SCI = 1); the authors give multiple real examples of AI mistranslating NL statements into Lean, producing mismatches; hence passing Lean verification offers no confidence in the original natural-language argument. The reposter quips this technically means gpt-6 repaired an incorrect Navier-Stokes proof.
More from Research
- New OpenAI paper extends Riemann zeta zero-free region to Re(s) > 7/8 — PTenigma · 2026-10-08
- OpenAI math paper on Weil classes reported flawed, raising doubts about unformalized proofs — ctjlewis · 2026-10-08
- ZooWork-ShopRanker: open e-commerce rerankers (0.6B-8B) aligned to shopping preferences — kalyan_kpl · 2026-10-08
- Ethereum's Justin Drake calls for 'bunker mode' crypto migration as AI math advances threaten assumptions — marcvanderchijs · 2026-10-08
- Lean proof may not map 1-to-1 to paper: GPT formalization takes shortcuts on hard lemmas — ctjlewis · 2026-10-08
- 'Alignment Whack-a-Mole' gets COLM 2026 oral slot, TV interview teased — TuhinChakr · 2026-10-08