DeepMind's AlphaProof Code verifies Signal security with 6.5M lines of Lean proofs at machine-code level
rohanpaul_ai · x · 2026-11-11
Google DeepMind unveiled AlphaProof Code, a multi-agent AI framework for formal specification and verification, and used it to prove key security properties of libsignal — Signal's reference implementation — at the machine-code level.
- The agents generated 6.5M lines of machine-checked Lean proof code resting on 26,000 lines of trusted specifications and formal AArch64 semantics; proofs are accepted only if Lean's kernel verifies them, and even the Rust compiler is not trusted.
- The proofs cover Signal's post-quantum PQXDH handshake and Triple Ratchet, establishing forward secrecy, authentication, post-compromise security, and handshake secrecy.
- It is the first time any method — with or without AI — has formally verified the actual production code of a widely deployed protocol at this depth, connecting protocol-level guarantees down to the instructions a processor actually runs.
- The team argues AI can make formal verification, historically too expensive for most software, practical for real-world systems. The Lean development is open at google-deepmind.github.io/code-verification.
More from Safety
- Congressional AI bill fight: federal preemption as 'floor, not a ceiling' — ShakeelHashim · 2026-11-12
- Standard binary hardening cuts Codex's 20 AI exploits down to just 2 — dawnsongtweets · 2026-11-12
- The patch-to-deployment window matters more than ever, CyberBinGym authors warn — dawnsongtweets · 2026-11-12
- Patches are clues: giving Codex the patched binary lifts exploit rate from 10% to 25% — dawnsongtweets · 2026-11-12
- CyberBinGym setup: 100 real CVEs, 53 projects, stripped binaries only — dawnsongtweets · 2026-11-12
- Dawn Song team's CyberBinGym: AI agents turn 1 in 5 binary patches into working exploits — dawnsongtweets · 2026-11-12