Axiom Math verifies Stacks Project’s 7,654 pages of math for AI agents
Axiom Math, a $1.6 billion startup, announced it has formally verified the Stacks Project—a 7,654-page mathematical library—into 16 million lines of Lean code. The project, started by Columbia mathematicians in 2005, is considered foundational in algebra, geometry, and number theory. Axiom’s AI agents translated and checked the proofs, ensuring logical consistency across interconnected theorems,…
Key points
- Axiom Math verified 7,654 pages of Stacks Project into 16 million Lean code lines
- Stacks Project is a 25-year-old, open-source math library with rare academic regard
- Axiom’s AI agents used parallel, self-directed tasks to preserve mathematical integrity
The startup, founded by CEO Carina Hong in 2025, raised $200 million in Series A funding last March. Axiom’s approach involves training custom AI models for math verification, a method Hong claims most competitors lack. The verified Stacks Project now serves as a reusable foundation for AI agents to build upon, addressing a gap in formalized algebraic geometry. Hong argues this could prevent ‘AI slop’ in software and scientific work by grounding future proofs in verified math.
Axiom Math wants to keep AI slop out of mathematics
TechCrunch AI · 5 November 2026
Loading the full article…
This text was published by TechCrunch AI and written by Aditya Mehta. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Researchers link AI agent power to human Goal attainment in formal models · 1 src
- PromptOps framework proposed for iterative LLM judge optimization · 1 src
- Researchers introduce BCSS to scale symbolic models of reasoning · 1 src
- Researchers test Cross-Agent formalization for clinical trial criteria · 1 src
- Researchers introduce efxplorer for Cooperative Multi-Agent reinforcement learning · 1 src
Comments
via GitHub Discussions