Gemini 智能体在 Lean 中生成超 600 万行证明，人工仅需复核关键 1%

摘要

Google DeepMind 称 Gemini 智能体在 Lean 中生成了超过 600 万行证明，并对代码进行审计，使人类专家只需复核其中关键的 1%。Google DeepMind 将其描述为用 AI 验证日常通信、电网等关键基础设施的一步，详情见 https://goo.gle/4BLrkbk。

摘要由模型根据原文撰写，以原文为准。

Gemini agents generated over 6M lines of proofs in Lean, auditing the code so human experts only had to review the critical 1%.

This is a key step toward using AI to verify vital infrastructure we all rely on – from daily communication to power grids. Find out more → https://goo.gle/4BLrkbk

GoogleDeepMind. Gemini agents generated over 6M lines of proofs in Lean, auditing the code so human experts only had to review the critical 1%.[EB/OL]. (2026-11-11)[访问日期]. https://x.com/GoogleDeepMind/status/2120484249321033911.

Gemini 智能体在 Lean 中生成超 600 万行证明，人工仅需复核关键 1%

摘要

正文 · 原文

引用原文