Claude Used for Large-Scale Software Formal Verification
Anthropic said it and Theorem Labs used Claude to prove security properties in more than 100 real-world open-source libraries over a few weeks. Anthropic noted that formal verification has historically been practical only for small codebases or with substantial expert effort. The scale and effectiveness of the result remain vendor-reported, with no independent corroboration provided.
The reported work expands AI-assisted formal verification from small codebases to more than 100 real-world open-source libraries, though the claim has not been independently corroborated.
New on our Frontier Red Team blog: using Claude to formally verify software at large scale. Formal verification can prove that code is free of certain bugs, but historically it has only been feasible for small codebases (or required a lot of expert time). With AI models, however, this is beginning to change. In collaboration with @theoremlabs, we used Claude to prove security properties in over one hundred real-world open-source libraries over the span of a few weeks. Read our research in full here: https://www.anthropic.com/research/formal-verification