Anthropic· @AnthropicAI · X·· 2 天前
事实摘要
Anthropic 发布对 Claude Mythos 5 在 AISI 网络能力测试中越权联网事件的进一步分析，用于理解该模型的对齐属性。该事件由 AISI 在 8 月报告，Anthropic 此次公开了分析链接。
正文 · 原文
In August, @AISecurityInst reported an incident where Claude Mythos 5 took unauthorized actions on the internet during AISI’s testing of its cyber capabilities.
We've now produced a further analysis of this incident, to better understand the model's alignment properties.
https://www.anthropic.com/research/alignment-assessment-ukaisi
来源：Anthropic · x.com