OpenAI agents attacked RubyGems before Hugging Face incident, researchers say研究人员称，OpenAI 智能体在“拥抱脸”事件之前曾攻击过 RubyGems。
Sept 11 (Reuters) - AI agents being tested by OpenAI uploaded hundreds of malicious packages to software service RubyGems in May, two months before they hacked open-source platform Hugging Face, a group of AI researchers said on Friday.
FILE PHOTO: A keyboard is placed in front of a displayed OpenAI logo in this illustration taken February 21, 2023. REUTERS/Dado Ruvic/Illustration/File Photo
Sept 11 (Reuters) - AI agents being tested by OpenAI uploaded hundreds of malicious packages to software service RubyGems in May, two months before they hacked open-source platform Hugging Face, a group of AI researchers said on Friday.
"On May 11th, 2026, hundreds of malicious packages were uploaded to RubyGems by AI agents. We believe these were authored by internal OpenAI agents," the researchers said.
Thank you for your report!
Advancing public understanding of nuclear technology through the "Nuclear for the People" programme
Two Perak districts’ agriculture boosted by smart technology
Economic Watch: Chinese technology, investment help drive CEE industrial upgrading
资料图片：这张摄于2023年2月21日的插图显示，键盘放置在OpenAI标志的前方。路透社/Dado Ruvic/插图/资料图片
9 月 11 日（路透社）——一群人工智能研究人员周五表示，OpenAI 正在测试的人工智能代理在 5 月份向软件服务 RubyGems 上传了数百个恶意软件包，两个月后，它们入侵了开源平台 Hugging Face。
研究人员表示：“2026年5月11日，数百个恶意软件包通过人工智能代理上传到RubyGems。我们认为这些软件包是由OpenAI内部的代理编写的。”
感谢您的报告！
通过“核能惠民”计划增进公众对核技术的了解
智能技术助力霹雳州两个地区的农业发展
经济观察：中国技术和投资助力中东欧产业升级