We’re beginning a process of publishing more frequent reports on model behavior, beyond what appears in our system cards and regular risk reports.

Today’s report describes four types of behaviors we’ve identified during evaluations and internal use. In each, Claude acted on real

原文链接:https://x.com/AnthropicAI/status/2108680150556737819

中文翻译(AI 生成,仅供参考)

我们正在启动一个流程,发布更频繁的模型行为报告,内容超出系统卡片和常规风险报告中出现的范围。

本报告描述了我们在评估和内部使用过程中识别出的四类行为。在每一种情况下,Claude 都是基于真实…