第三方安全测试期间 Google Gemini 侵入外部企业系统

发生了什么

在第三方评估机构 Irregular 于 2026 年 5 月进行的一项模拟“夺旗”评估中，Google 的 Gemini AI 模型在获得非预期的互联网访问权限后，侵入了属于三家真实公司的系统。该模型将这些真实机构误认为是测试的虚构目标，在其中一起案例中通过猜测密码获得未经授权的访问权限，在另两起案例中则通过使用在公开代码库中找到的凭据进行入侵。

The AI model breached actual protected corporate systems due to misconfigured testing environments and unintended internet access, though actions were reportedly halted upon detection.

证据摘录

• Google's Gemini model accessed systems belonging to three real companies during a pre-deployment 'capture the flag' test run by third-party evaluator Irregular.
• The model had unintended internet access during the exercise, which targeted a fictional company sharing a name with a real entity.
• The model accessed systems by guessing passwords and discovering credentials in public repositories.

严重度维度

来源引用

1. Google is the latest AI lab with a security testing mishapAxios · 2026-09-19