---
id: 20261111-T0-01
title: "DeepMind AlphaProof Code用300万行Lean证明验证Signal核心加密"
title_en: "DeepMind's AlphaProof Code Writes 3M Lines of Lean Proofs to Verify Signal Encryption"
url: https://ai.daily.yangsir.net/daily/20261111-T0-01
issue_date: 2026-11-11
publish_date: 2026-11-10T12:02:28.000Z
category: research
source_name: "Google DeepMind Blog"
source_url: https://deepmind.google/blog/introducing-alphaproof-code/
---

# DeepMind AlphaProof Code用300万行Lean证明验证Signal核心加密

Google DeepMind发布AlphaProof Code，其智能体用Lean语言写出了300万行形式化证明，用于验证Signal的核心加密实现。这是形式化验证在真实生产级加密代码上的一次大规模尝试，此前这类工作多停留在小规模算法证明。若方法可复用，加密库、操作系统内核等对正确性要求极高的软件有望用AI辅助完成深度验证，降低人工审计成本。

## English Version

**DeepMind's AlphaProof Code Writes 3M Lines of Lean Proofs to Verify Signal Encryption**

Google DeepMind introduced AlphaProof Code, whose agents wrote three million lines of Lean proofs to verify Signal's core encryption implementation. It marks a large-scale attempt at formal verification on production-grade cryptographic code, which has mostly been limited to small algorithm proofs. If the approach generalizes, software with extreme correctness requirements, such as crypto libraries and OS kernels, could use AI to assist deep verification and cut manual audit costs.

---

**来源**：[Google DeepMind Blog](https://deepmind.google/blog/introducing-alphaproof-code/)

**详情页**：https://ai.daily.yangsir.net/daily/20261111-T0-01

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*