199 reports of soundness bugs in proof checkers collected on one page
jedisct1 · x · 2026-11-13
Cryptographer David Wong (hashbreaker) posted a new page, "False computer-checked proofs," compiling 199 reports of soundness bugs across major proof-checking programs including Coq/Rocq, Isabelle, Lean, HOL, Metamath and Mizar. An accompanying chart shows Coq/Rocq-related reports accelerating sharply through 2025-2026, far outpacing other assistants. The takeaway: a computer-checked proof is only as trustworthy as the kernel checking it, and these bugs mean some formally "verified" results may be unsound — a significant warning for security and mathematics communities relying on formal verification.
More from Safety
- AI safety org ML4Good gets $7.4M; 186 alumni now work in the field — HaydnBelfield · 2026-11-13
- Robert Wiblin puts 80% death risk on max-speed ASI, 30% with 3-year delay — trevposts · 2026-11-13
- AI safety discourse ignores 60%+ of potential harms, critic says — 90% is technical safety — amplifiedamp · 2026-11-13
- Ad tool Punk targets phones that visited any location, bypassing Meta's algorithm — JaynitMakwana · 2026-11-13
- Apify confirms breach: unauthorized party accessed user account info Nov 9-10 — gaganghotra_ · 2026-11-13
- NGOs sue OpenAI, Anthropic, Google, Meta and SpaceX to freeze advanced AI development — GarrisonLovely · 2026-11-13