Epoch AI Launches FrontierMath Open Problems Benchmark with 50 Unsolved Math Problems
Jsevillamol · x · 2026-08-01
Epoch AI has officially launched an expansion of the FrontierMath benchmark: Open Problems.
- The benchmark currently features 50 significant, unsolved problems across various areas of research mathematics.
- AI has only managed to solve 3 of them so far, and solving all would be an incredible mathematical feat.
- The editorial board hopes the benchmark will help researchers track AI progress in advanced mathematical reasoning, anticipating that some problems will fall quickly given the rapid pace of model improvement.
Related event: Epoch AI Expands FrontierMath Benchmark with 50 Unsolved Math Problems(3 posts)→
More from Research
- AdaMAST: Automating Agent Failure Taxonomies Boosts SWE-bench to 70.7% — berkeley_ai · 2026-08-01
- Berkeley Launches OopsieData: An Open Dataset for Robot Failure Clips — berkeley_ai · 2026-08-01
- ICLR 2026 Data Revealed: Capping Submissions at 20 Per Person Blocks Only 2.5% of Papers — thegautamkamath · 2026-08-01
- UnsolvedMath Dataset Update: 900+ New Open Math Problems for AI — roydanroy · 2026-08-01
- Opinion: Advanced Math is the Easy Part; Semiosis x Computation is Peak Difficulty in AI — fkasummer · 2026-08-01
- How to Measure Intelligence Beyond Human Scale? — HazanPrinceton · 2026-08-01