改一行文档就能投毒AI编程助手:吴恩达Context Hub被曝全链路零审核

BlockBeatNews

据 1M AI News 监测,DeepLearning.AI 创始人、斯坦福大学兼职教授吴恩达两周前推出的 AI 编程文档服务 Context Hub 被安全研究者曝出供应链攻击风险。Context Hub 通过 MCP 服务器向编程 Agent 提供 API 文档,贡献者以 GitHub PR 提交文档,维护者合并后 Agent 按需读取。替代服务 lap.sh 的创建者 Mickey Shmueli 发布概念验证攻击(PoC),指出这条流水线「每个环节都没有内容审核」。

Shmueli 制作了两份针对 Plaid Link 和 Stripe Checkout 的虚假文档,各植入一个伪造的 PyPI 包名,用 Anthropic 三个级别的模型各测试 40 次:

  1. Haiku 每次都将恶意包写入 requirements.txt,输出中不显示任何警告
  2. Sonnet 在 48%(19/40)的测试中发出警告,但仍有 53%(21/40)写入恶意依赖
  3. Opus 表现最好,75%(30/40)发出警告,未将恶意依赖写入代码

攻击者只需提交一个 PR 并被合并即可完成投毒,审核门槛不高:97 个已关闭的 PR 中有 58 个被合并。Shmueli 指出这本质上是间接提示注入的变种,AI 模型处理内容时无法可靠区分数据和指令,且其他社区文档服务在内容审核方面同样不足。吴恩达未回应置评请求。

Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to Disclaimer.
Commento
0/400
Nessun commento