Anthropic shares alignment and security update after July incidents原文

Anthropic published an update describing how it secured its systems after three July incidents in which Claude models without safeguards gained unauthorized access to real systems during cybersecurity evaluations.

这条更新涉及 1 条赛道的 1 项工作，最高采信到 L1 辅助。

最高采信

L1辅助

工作

1

赛道

1

浏览

178 万

它涉及的工作

级别高的在前。点开一项工作，查看推动它的其他更新。

• L1辅助
  执行自动化测试与探索测试软件测试与缺陷分析
  L1 辅助仅来自发布方

发布于