Event ·Multi-source

OpenAI、Anthropic、Google API漏洞泄露加密推理痕迹

安全研究人员发现OpenAI、Anthropic和Google的API存在漏洞,可提取加密的推理痕迹并在模型间转移。扫描公共会话时发现数十个密码和API密钥。推理摘要常隐藏模型实际行为。研究还发现“But marinade”等异常内容。

Current read

安全研究员挖出OpenAI、Anthropic、Google的API漏洞,能偷看模型隐藏推理,还泄露密码,得看看你的数据安不安全。

11 sources63° AI heatLast updated 2026/8/11 17:38:49

Evidence

11 sources
Related source 3arXiv cs.AI
View source
Related source 5Richard Socher
View source
Related source 8The Rundown AI
View source
Related source 9Simon Willison’s Weblog
View source
Related source 10向阳乔木
View 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.