What happened

Anthropic used Claude to mathematically prove that code is correct. The model produced roughly 180 million lines of proof code for 123 open-source libraries.

Why it matters

Formal verification proves that software behaves correctly under defined conditions using mathematical proofs rather than tests. It is normally labor-intensive and therefore rare. Anthropic says Claude automated the process and caught bugs that years of testing had missed.

The goal

The company aims to make AI-driven formal verification as routine as standard testing, which could speed up bug hunting in security-critical software.