# 研究者用 Claude Opus 5 将图片上传漏洞变成 OpenAI 员工账号接管

- 来源：Yuchen Jin (@Yuchenj_UW)
- 发布时间：2026-09-18 10:48
- AIHOT 分数：62
- AIHOT 链接：https://aihot.news/items/cmu6dh1bx0kbqrofj2puz2q3i
- 原文链接：https://x.com/Yuchenj_UW/status/2100778872728060304

## AI 摘要

3 名安全研究者用 Claude Opus 5 将一个图片上传漏洞升级为 OpenAI 员工账号接管，并让被接管的员工账号通过 Codex 在 OpenAI 内部 monorepo 开了一个 PR，整个攻击的 token 成本不足 3000 美元。作者称 Opus 4.8 难以完成该 exploit，Opus 5 发布后在数小时内破解，并认为最好的防御是把最强的 AI 交给防御者。

## 正文

OK, this is a big deal:

3 researchers used Claude Opus 5 to turn an image upload bug into an OpenAI employee account takeover, then had the compromised employee’s Codex open a PR in OpenAI’s internal monorepo.

Their entire hacking cost less than $3000 in tokens.

Opus 4.8 struggled with the exploit. Then Opus 5 dropped and cracked it within hours.

AI-powered cyberattacks are becoming common and cheap. The best defense is to put the best AI in the hands of defenders too.