Axiom Math, a $1.6 billion startup, announced it formally verified the Stacks Project, one of the largest bodies of mathematics ever written by humans. The company's AI agents translated the Stacks Project's roughly 7,654 pages into about 16 million lines of computer-checked code written in Lean, a programming language used for formal verification.
Formal verification rewrites proofs as code and uses a proof assistant to check each line. If the proof passes, mathematicians can trust it is logically sound. The interconnected nature of the Stacks Project, where theorems build on one another, made the verification challenging. Axiom's team had to keep every connection intact across the 16 million lines of code.
CEO Carina Hong, 25, dropped out of a Stanford math PhD program and founded Axiom in March 2025. In March, the company raised a $200 million Series A led by Menlo Ventures.
The verification fills what Ashvin Swaminathan, a member of Axiom's technical staff, called "this huge black hole" in formalized mathematics. That gap covers algebraic geometry, an area where researchers previously had little verified foundation to build on with Lean.
Axiom trained its own math-focused models rather than relying on existing AI systems, an approach Hong said most other verification startups have not taken. The team gave its agents a loose structure and let them work in parallel, identifying tasks for themselves.
Beyond mathematicians, Axiom targets software developers who struggle to guarantee their code is free of bugs and security holes. The company argues that as AI agents write more code, software will need to rest on verified foundations. Hong said verified results become building blocks so future proofs can build on checked math instead of starting from scratch.