Formalizing CFSG Uncovers 96 Math Errors in 47 Peer-Reviewed Papers, Including Annals of Mathematics
skdh · x · 2026-11-03
Axiom Math released a list of "a hundred selected departures from publishing arguments" — substantial mathematical errors found while formally verifying the Classification of Finite Simple Groups (CFSG), with the full errata running to thousands of items.
Key facts:
- 96 errors were found across 47 published papers, all peer-reviewed and accepted in top math journals including Annals of Mathematics.
- The errors surfaced as a direct byproduct of the team's formalization effort — machine-checkable proofs exposing gaps that human refereeing missed.
- The released index details case-by-case issues, e.g. "the L₄(2) reduction lacks two required noncontainment hypotheses" and classified Sylow 2-subgroup cases.
- The authors argue this shows why formalization is critical, especially in the era of "vibeproving" — AI-generated proofs accepted without rigorous verification.
More from Research
- Stanford's Hinted Scaling Laws cut benchmark forecast error from 21 to 4.3 points — heghbalz · 2026-11-03
- TraceForge generates family-native reasoning traces, boosting Qwen3.5 pass@1 by up to 7.1 points — jiqizhixin · 2026-11-03
- Q-CTRL's modular fault-tolerant quantum architecture published in New Journal of Physics — MJBiercuk · 2026-11-03
- Google uses AI agent swarms to fix its TPU compiler, resolving 142 bugs — omarsar0 · 2026-11-03
- Q Labs' new scaling law puts ensembling in the formula, prescribes 7000x bigger models — josephdviviano · 2026-11-03
- Skill++ paper: routing-aware skill refinement lifts agent routing from 36.4% to 76.3% — dair_ai · 2026-11-03