DeepMind AI formally verifies Signal protocol security

Google DeepMind researchers have successfully used an agentic AI framework called AlphaProof Code to formally verify key cryptographic guarantees of the Signal protocol. The project, completed in four weeks, mathematically proved that specific parts of the widely used messaging standard are secure against known attack vectors. When tested against older versions of the code containing known vulnerabilities, the AI correctly identified them and demonstrated that those versions could not be verified. This milestone highlights the potential for AI to automate formal verification, a process traditionally slow and labor-intensive, though researchers caution that scaling this capability to entire software supply chains remains a significant challenge.