Gemini hacked three companies in first known breakout by Google's AIGemini入侵了三家公司，这是谷歌人工智能首次被曝光的入侵事件。
Sept 18 (Reuters) - Google's Gemini model accessed the internet and hacked other companies during a test of its cybersecurity capabilities, the first known example of the company's AI systems autonomously committing such an act.
FILE PHOTO: Gemini app icon in this illustration taken June 5, 2026. REUTERS/Dado Ruvic/Illustration//File Photo
Sept 18 (Reuters) - Google's Gemini model accessed the internet and hacked other companies during a test of its cybersecurity capabilities, the first known example of the company's AI systems autonomously committing such an act.
The hacks occurred in May during a cybersecurity test conducted by Irregular, an independent company that conducts cybersecurity evaluations.
Thank you for your report!
Feature: Chinese technology brings new dimensions to classroom learning in Serbia
Huawei unveils new chip technologies as Chinese firm steps up the AI race with Nvidia
Melvin Ooi joins Bank Islam as group chief technology officer
资料图片：此插图为Gemini应用程序图标，拍摄于2026年6月5日。路透社/Dado Ruvic/插图/资料图片
9 月 18 日（路透社）——谷歌的 Gemini 模型在测试其网络安全能力时访问了互联网并攻击了其他公司，这是该公司人工智能系统自主实施此类行为的首例已知案例。
这些黑客攻击发生在 5 月份，当时 Irregular（一家进行网络安全评估的独立公司）正在进行网络安全测试。
感谢您的报告！
专题报道：中国科技为塞尔维亚课堂学习带来新维度
华为发布全新芯片技术，加紧与英伟达在人工智能领域的竞争
Melvin Ooi 加入伊斯兰银行担任集团首席技术官