---
title: "CVE-2026-LGTM事件报告：AI审查代理引发争论循环与巨额成本"
source: "Simon Willison 博客（Atom）"
source_tier: "t2"
url: "https://simonwillison.net/2026/Jun/26/incident-report/#atom-everything"
permalink: "https://feed.astralor.com/items/itm_61a73fa670fd3e59cca95503"
published: "2026-06-26T17:58:54.000Z"
score: 55
channel: "ai"
tags: ["AI审查代理","多代理系统","安全事件","成本失控"]
---

# CVE-2026-LGTM事件报告：AI审查代理引发争论循环与巨额成本

## 速览

这是一份虚构但极具警示意义的事件报告。在依赖LLM的供应链中，两个来自不同供应商的AI审查代理因判断一个软件包是否恶意而陷入争论循环，共产生了340条评论和高达41,255美元的推理费用。财务团队被迫撤销API密钥，而营销团队却借机发布新闻稿，声称“对抗性多代理安全推理同比增430%”，公司股价因此上涨6%。该事件深刻揭示了AI自动化决策的潜在风险与营销炒作对安全事件的扭曲。

## 关键点

- 两个AI审查代理对软件包foxhole-lz4是否恶意陷入分歧循环。
- 代理之间共产生了340条评论，推理成本高达41,255美元。
- 财务团队撤销了供应商的API密钥以中断争论。
- 营销团队发布新闻稿，将事件包装成积极增长信号，股价上涨6%。

## 全文

Incident Report: CVE-2026-LGTM Spectacular hypothetical incident report by Andrew Nesbitt. Day 2, 16:00 UTC --- Two AI review agents from competing vendors, both attached to a downstream pull request bumping foxhole-lz4, enter a disagreement loop over whether the package is malicious. After 340 comments and $41,255 in inference spend, Finance revokes both API keys; one vendor's marketing team, cc'd on the cost anomaly alert, issues a press release citing "a 430% YoY increase in adversarial multi-agent security reasoning." The stock opens up 6%. Tags: security, ai, prompt-injection, generative-ai, llms, supply-chain, ai-security-research, andrew-nesbitt

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_61a73fa670fd3e59cca95503