AlphaProof Code verifies Signal's cryptographic core in four weeks

Google DeepMind has unveiled AlphaProof Code, an AI system that mathematically proves software meets its security specifications without errors. Researchers demonstrated it on Signal's encryption protocol, completing in four weeks work that typically takes years. The system verified the actual cryptographic core code running on users' devices, not just the theoretical design.

How the verification worked

After experts defined formal security requirements, AI agents built on Gemini Carbon wrote more than six million lines in the Lean proof language. They flagged roughly one percent of the codebase for human review. Two bugs planted deliberately for testing were caught immediately. AI agents in AlphaProof Code also help create the specifications and then independently prove that the program code meets them.

Broader implications

Earlier verification efforts stopped at the theoretical design stage, according to a blog post by DeepMind researchers Pushmeet Kohli and Swarat Chaudhuri. Signal CTO Ehren Kret said the approach strengthens trust in data privacy as AI threats grow. DeepMind plans to extend the method to power grids and medical devices through systems like CodeMender.

Sources: The Decoder