Google DeepMind AI tool formally verifies key Signal protocol guarantees

The short version

• Google DeepMind researchers used the AlphaProof Code agentic AI framework to formally verify key cryptographic guarantees of the Signal protocol, according to Wired.[Wired]
• The four-week proof of concept covered only portions of the protocol, not the full Signal mobile app, and found no new vulnerabilities in those portions.[Wired]
• The analysis correctly flagged known, long-fixed flaws in older Signal versions, and DeepMind is releasing its specification and proof to aid other verification work.[Wired]
• Google declined to disclose the analysis cost or say whether it will offer a commercial product, and Wired reports that DeepMind's Swarat Chaudhuri said scaling to all software remains a long way off.[Wired]

Key facts

• Google DeepMind researchers used agentic AI to formally verify key cryptographic guarantees of the Signal protocol.[Wired]
• The work was a proof-of-concept project that took four weeks and used a new AI framework called AlphaProof Code.[Wired]
• The analysis did not formally verify the entire Signal mobile app.[Wired]
• The analysis found no new vulnerabilities in the portions of the protocol it examined.[Wired]
• When applied to older Signal versions with known, long-fixed vulnerabilities, the tool correctly identified them.[Wired]
• The researchers are releasing the specification and proof they developed to aid other formal verification work.[Wired]
• Google conducted the analysis free of charge as a research project and declined to share its cost.[Wired]

Sources

Outlet counts describe coverage, not independent confirmation. Reports may share a wire service or original source.

• A New AI Tool From Google DeepMind Proved the Signal Protocol Is SecureWired metered
• AlphaProof CodeHacker News
• Google DeepMind analysis proves Signal and WhatsApp encryption is secure9to5Mac