Event ·Multi-source

研究人员用 Claude 洞悉 OpenAI 算法秘密

三名安全研究员用 Anthropic 的 Claude Opus 4.8 和 5 模型，在72小时内成功入侵 OpenAI 员工账户，获取了其 GitHub 代码库“Monorepo”的访问权限，该库据称包含 OpenAI 的核心算法信息。

Current read

朋友发现个好玩的事，三个安全研究员用 Claude 模型，花不到三天就搞定了 OpenAI 的员工账户，还拿到了他们的代码库，挺有意思的。

11 sources58° AI heatLast updated 2026/9/18 15:30:16

Evidence

11 sources

Primary sourcetheverge

2026/9/18 15:30:16View source

Related source 1Decoder

2026/9/18 17:20:30View source

Related source 2lmarena.ai

2026/9/16 19:13:56View source

Related source 3IT之家

2026/9/18 05:26:43View source

Related source 4techcrunch

2026/9/18 14:00:14View source

Related source 5宝玉

2026/9/18 04:57:51View source

Related source 6Gary Marcus

2026/9/18 21:58:14View source

Related source 7AI Will

2026/9/20 08:54:06View source

Related source 8Genk

2026/9/20 11:01:00View source

Related source 9Justine Moore

2026/9/16 23:48:07View source

Related source 10arXiv: OpenAI

2026/9/17 02:14:53View 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