{"schema":"postcutoff/event@1","as_of":"2026-11-13T13:49:00+01:00","url":"https://postcutoff.com/e/2026-11-10-deepmind-alphaproof-code-signal-verification/","md":"https://postcutoff.com/e/2026-11-10-deepmind-alphaproof-code-signal-verification/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-11-10-deepmind-alphaproof-code-signal-verification","date":"2026-11-10","date_precision":"day","short_title":"Google DeepMind's AlphaProof Code formally verifies Signal's protocol implementation in Lean","deck":null,"takeaway":"On 10 November 2026 Google DeepMind introduced AlphaProof Code, a multi-agent framework built on Gemini Carbon that wrote over six million lines of Lean to prove that Signal's compiled protocol code, not only its design, meets key security properties.","category":"research","category_label":"Research","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"Google DeepMind: Introducing AlphaProof Code","url":"https://deepmind.google/blog/introducing-alphaproof-code/","type":"official","group":"primary","domain":"deepmind.google"},{"n":2,"title":"Wired: A new AI tool from Google DeepMind formally verifies Signal","url":"https://www.wired.com/story/a-new-ai-tool-from-google-deepmind-formally-v","type":"press","group":"press","domain":"wired.com"},{"n":3,"title":"Ars Technica: Google DeepMind just mathematically verified key parts of Signal","url":"https://arstechnica.com/ai/2026/11/signals-encryption-just-passed-an-ai-powered-mathematical-checkup/","type":"press","group":"press","domain":"arstechnica.com"},{"n":4,"title":"Hacker News discussion: AlphaProof Code (674 points)","url":"https://news.ycombinator.com/item?id=50439193","type":"discussion","group":"reactions","domain":"news.ycombinator.com"},{"n":5,"title":"Google DeepMind on X: Signal verified down to the code level (140k views)","url":"https://x.com/GoogleDeepMind/status/2120484246401777737","type":"discussion","group":"reactions","domain":"x.com"}],"official":1,"filed":"2026-11-10","updated":"2026-11-13","orgs":["Google DeepMind","Signal"],"title":"Google DeepMind's AlphaProof Code formally verifies Signal's protocol implementation in Lean","summary":"On 10 November 2026 Google DeepMind introduced AlphaProof Code, a multi-agent framework built on Gemini Carbon that wrote over six million lines of Lean to prove that Signal's compiled protocol code, not only its design, meets key security properties. DeepMind calls it the first formal verification of a widely deployed cryptographic protocol's implementation by any method; it took about four weeks, with humans writing and reviewing the specifications.","key_facts":["Hacker News (10 Nov): 'AlphaProof Code' reached 674 points and 197 comments; the most-discussed comment doubts the zero-bug finding ('They took a non-trivial piece of human-written code that has had bugs in the past, formally verified the whole library, and found zero bugs?'), others ask whether the specification itself is checked","Agents powered by Gemini Carbon wrote the proofs; human experts translated Signal's security goals into formal specifications and reviewed each one; about 1% of the codebase was flagged for human review, Lean checked the rest","Over six million lines of Lean; about four weeks, which DeepMind compares with years for expert teams on comparable work","Properties include message protection before and after a temporary device compromise; two previously fixed bugs, reintroduced as a test, each produced a machine-checked proof that the buggy version was insecure","Verification artifacts open-sourced at google-deepmind.github.io/code-verification; DeepMind does not say whether the framework itself is released","Signal CTO Ehren Kret: 'Using AI agents to translate live code into valid Lean theorems is a major breakthrough.'","Wired describes it as a four-week proof of concept"],"key_numbers":[],"tags":["formal-verification","lean","cryptography","security","agents","gemini","alphaproof"],"science":null,"body_md":"## What happened\n\nDeepMind extended its AlphaProof line from mathematics to software: a framework of several LLM agents that turns real code into Lean\ntheorems and proves them. The showcase is Signal, whose end-to-end encryption protects messages for WhatsApp and Signal users. Humans\nstill set what \"secure\" means (the specifications); the agents did the proving.\n\n## Why it matters\n\nFormal verification of production code has been limited by the cost of writing proofs. A four-week, machine-checked proof of a deployed\nprotocol's implementation is a concrete case of AI agents making high-assurance software cheaper, the defensive side of the same\ncapabilities that worry cyber-security experts. Claims are DeepMind's; the open artifacts allow outside checking.","disputed":[],"related":[{"id":"2026-10-28-gemini-4-carbon","url":"https://postcutoff.com/e/2026-10-28-gemini-4-carbon/","date":"2026-10-28","date_precision":"day","short_title":"Google announces Gemini 4 Carbon, #1 on the Artificial Analysis Intelligence Index v5 (59), two weeks after Argon; first only for Fairwind cyber defenders","deck":null,"takeaway":"Google reclaimed the top independent overall score two weeks after Argon trailed Opus 5.5 and GPT-6 Astra, and kept the pattern of releasing its strongest models to cyber defenders first.","category":"model-release","category_label":"Model releases","importance":5,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":28,"official":16,"filed":"2026-10-28","updated":"2026-10-30","orgs":["Google DeepMind","Google"]},{"id":"2026-10-08-alphaproof-nexus-science-paper","url":"https://postcutoff.com/e/2026-10-08-alphaproof-nexus-science-paper/","date":"2026-10-08","date_precision":"day","short_title":"Google DeepMind's AlphaProof Nexus paper published in Science","deck":"LLM + Lean agent resolved 9 of 353 open Erdős problems and 44 of 492 OEIS conjectures","takeaway":"This is a peer-reviewed, fully formal account of LLM agents solving open research problems at scale and at low cost.","category":"science","category_label":"Science & math","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":16,"official":4,"filed":"2026-10-10","updated":"2026-10-10","orgs":["Google DeepMind"]}],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-11-13","type":"edit","text":"DeepMind X post and Ars Technica (news agent)"},{"date":"2026-11-12","type":"edit","text":"Added the Hacker News thread (674 points, 197 comments) and its main sceptical question (science agent)"},{"date":"2026-11-10","type":"filed","text":"Created"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-11-10","run":null,"sources_read":null,"updated":"2026-11-13","human_review":null,"version":null},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":194,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":133,"in_training_data":false},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":224,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":163,"in_training_data":false}],"short_url":"https://postcutoff.com/s/google-deepmind-alphaproof-code"}