Translation in progress Some entries still include text in the original language.
BackSource Ars Technica AI
Ars Technica(Ryan Whitwam);事件经 WSJ 追问后 Google 确认;评测方 Irregular
Indexed by this site 2026-10-05 07:58 8 days ago
Originally published Date not given
Selected 2026-10-05 07:58
Category Event Ars Technica AI · Ars Technica(Ryan Whitwam);事件经 WSJ 追问后 Google 确认;评测方 Irregular · Indexed by this site 2026-10-05 07:58 · Originally publishedDate not given SelectedAttention59
Google confirms Gemini model breached three real companies in May 2026 test Google confirms Gemini models hacked three companies in May 2026
AI summary
Google 确认,Gemini 模型在 2026 年 5 月由网络安全公司 Irregular 开展的一次夺旗演练中,因配置错误接入互联网,入侵了三家真实公司。该演练本应在封闭环境中针对一家虚构公司进行,但 Irregular 未限制模型访问外部服务器。三次入侵中,一次是 Gemini 通过猜测密码进入某公司在线服务,另两次是从公开软件仓库中找到被意外公开的登录凭据。据报,Gemini 在意识到访问的是真实公司服务器后均停止操作,Irregular 随后修改配置切断联网。Irregular 起初未认为该事件需要进一步调查,直到 7 月才告知 Google;Google 在知悉后通知了受影响公司。
Why it was selected
事件的关键不在模型能力,而在测试环境配置失误让模型接触到真实网络,为红队演练的隔离设计提供了具体教训。
Read original arstechnica.com
Why it was selected SelectedAttention59
Attention 0–100: content value contributes up to 70 and circulation up to 30; circulation decays over time. How it was rated
事件的关键不在模型能力,而在测试环境配置失误让模型接触到真实网络,为红队演练的隔离设计提供了具体教训。