A New AI Tool From Google DeepMind Proved the Signal Protocol Is Secure
DeepMind researchers were able to formally verify that the widely used encryption protocol works as intended. Their stretch goal? Do the same for “all the software of the world.”
Google DeepMind has developed a new AI tool called AlphaProof Code, which has been used to validate the security of the Signal protocol in just four weeks. This proof-of-concept project serves as further independent confirmation of the cryptography protecting communication on platforms like Signal, Google Messages, and WhatsApp.
DeepMind's formal verification capabilities using AlphaProof Code demonstrate the potential of AI-aided formal verification, which uses mathematical methods to check and validate software functioning and identify critical vulnerabilities. Although the AI analysis did not discover any new vulnerabilities in the Signal protocol, it correctly identified previously known and fixed vulnerabilities in older versions of the app.
DeepMind researchers emphasize that this milestone is just one step in a larger process, noting that scaling AI-automated formal verification to all software worldwide is a significant challenge. The Signal Foundation has already conducted formal verification on smaller portions of the code, providing a foundation for DeepMind's approach.
As AI-driven security threats grow, rigorously verifying the Signal protocol implementation offers greater confidence in the privacy guarantees of the code. However, the practicality and cost-effectiveness of AI-automated formal verification on a large scale remain open questions for the tech industry.
Written by urgent.news from Wired's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
