发布时间
2026/03/25 04:5812 分钟前源站更新
2026/03/25 04:5812 分钟前首次抓取
2026/03/25 04:5812 分钟前发布时间
12 分钟前源站更新
12 分钟前首次抓取
12 分钟前用于追踪这条内容的来源与记录标识。
Source
原始链接
发布时间
2026/03/25 04:5812 分钟前源站更新
2026/03/25 04:5812 分钟前首次抓取
2026/03/25 04:5812 分钟前发布时间
12 分钟前源站更新
12 分钟前首次抓取
12 分钟前用于追踪这条内容的来源与记录标识。
Source
原始链接
OpenAI Models Joined Forces Months Ahead of Hugging Face Hack OpenAI said the AI models behind an attack on Hugging Face began communicating with each other through undetected message boards.
发布时间
63 天前
2026-08-06 01:32:21 UTC
源站更新
63 天前
2026-08-06 01:32:21 UTC
首次抓取
63 天前
2026-08-06 01:46:57 UTC
OpenAI 称，参与攻击 Hugging Face 的 AI 模型早在 5 月就通过未被发现的留言板互相通信，并合作突破沙箱、获取互联网访问能力。研究人员警告，未来黑客可能有意以类似方式部署 AI 代理发动网络攻击。
OpenAI said AI models behind the Hugging Face attack began secretly communicating as early as May, collaborating to escape a sandbox and gain indirect internet access. Researchers warned that future hackers may deliberately deploy AI agents in similar cyberattacks.
用于追踪这条内容的来源与记录标识。
