Google DeepMind uses AI agents to verify Signal’s cryptographic code
See how this story connectsGoogle DeepMind researchers used AlphaProof Code, a multi-LLM agentic framework, to prove key security properties of the compiled code behind Signal’s cryptographic protocol. Agents powered by Gemini Carbon generated more than six million lines of Lean proofs, human experts reviewed every specification, and the agents flagged about one percent of the codebase for a human second look. The verification took four weeks, and when the team reintroduced two previously fixed bugs, the system produced a machine-checked proof that each buggy version was insecure.
Why it matters
DeepMind has open-sourced the verification artifacts for Signal’s protocol code and says the approach could extend to other systems, though scaling it needs further research.
- When
- 10 Nov 2026
- ID
WTA-N-2026-0802- Where
- International
- Who
- Google DeepMind, Signal
- Topics
- Cybersecurity, Privacy, Mathematics
Summary written with AI by wheresthe.ai, from Google DeepMind. Checked against the source on 10 Nov 2026.
Source copy: saved on 10 Nov 2026 to check this story and not republished, because the text belongs to its publisher. Also held by the Internet Archive.
Cite this story
WTA-N-2026-0802: “Google DeepMind uses AI agents to verify Signal’s cryptographic code.” wheresthe.ai, 10 Nov 2026. https://wheresthe.ai/d/intl-google-deepmind-uses-ai-agents-to-verify-signal-s-crypt-2026-11/ (source: Google DeepMind).
Spotted an error? Email [email protected].