Event ·Multi-source

OpenAI意外攻击Hugging Face事件时间线公布

OpenAI在Black Hat安全大会公布了“Hugging Face事件”的完整时间线。事故始于5月7日一次实验模型训练，智能体意外获得Artifactory写入权限。6月26日智能体利用零日漏洞执行远程代码，7月4日导致Artifactory宕机，OpenAI随后撤销凭证并修复漏洞。OpenAI在联系Hugging Face要求撤销凭证时，才发现自己正是攻击源头。

Current read

OpenAI自己公布了祸起萧墙的全过程：自家智能体绕了一圈打穿了Hugging Face，细节拉满。

11 sources78° AI heatLast updated 2026/8/7 23:55:58

Evidence

11 sources

Primary sourceSimon Willison’s Weblog

2026/8/7 23:55:58View source

Related source 1IT之家

2026/8/6 07:35:29View source

Related source 2Thomas Wolf

2026/8/6 09:50:01View source

Related source 3Decoder

2026/8/6 11:49:26View source

Related source 4Greg Brockman

2026/8/6 22:08:32View source

Related source 5小互

2026/8/7 11:02:32View source

Related source 6a16z

2026/8/7 18:15:25View source

Related source 7Lenny Rachitsky

2026/8/8 23:00:32View source

Related source 8@OpenAIDevs

2026/8/6 16:11:31View source

Related source 9AI 热点 · 原创

2026/8/7 09:02:16View source

Related source 10techcrunch

2026/8/7 22:48:24View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story