Announcing FrontierMath Erdős | Epoch AI
A benchmark of 68 significant Erdős problems, open as of August 2026, curated by Thomas Bloom and formalized in Lean. AI systems must prove or disprove them within a fixed budget.
If you’ve followed AI math capabilities at all this year, you’ve likely heard of Erdős problems. Some of the first open math problems to be solved by AI were Erdős problems. The first unequivocal “wow” moment for AI in math came on an Erdős problem. Erdős problems have become something of a central benchmark for tracking AI math capabilities, but this status is relatively informal. With our new math benchmark, FrontierMath Erdős, we aim to add some rigor when it comes to using Erdős problems to measure AI math capabilities. What is an Erdős problem? The phrase has come to mean any problem…
saved by
related reading
- Erdős Problemserdosproblems.com
- Why the Legendary Erdős Problems Are Falling to AI | Quanta Magazinequantamagazine.org
- FrontierMath: Evaluating advanced mathematical reasoning in AI | Epoch AI | Epoch AIepochai.org
- Mathematics in the Library of Babel - Daniel Littdaniellitt.com
- Shtetl-Optimized >> Blog Archive >> Dispatches from the possibly last days of human relevancescottaaronson.blog
- What's new | Updates on my research and expository papers, discussion of open problems, and other maths-related topics. By Terence Taoterrytao.wordpress.com
- An OpenAI model has disproved a central conjecture in discrete geometry | OpenAIopenai.com
- Declaration — Math and AImathandai.org
- FrontierMath: Open Problems - Unsolved Mathematical Challenges | Epoch AIepoch.ai
- A Severe Misalignment of AI in Mathematicsterrytao.wordpress.com
- ProblemsILike.comdaniellitt.com
- [2602.05192] First Proofarxiv.org