Anthropic 发布文章,就安全评测的设计、运行与报告给出 10 条建议
AI 导读
Anthropic 在其工程博客发布文章,就如何设计、运行和报告安全评测给出 10 条建议。文章称,安全评测对判断模型是否可部署、应构建哪些护栏、训练与研究应聚焦何处十分关键。原文链接:https://www.anthropic.com/research/building-strong-safety-evaluations
正文 · 原文
Safety evaluations are critical for understanding whether a model is ready to deploy, which safeguards to build for it, and where to focus training and research.
On our Engineering Blog, we offer 10 recommendations for how to design, run and report them.
https://www.anthropic.com/research/building-strong-safety-evaluations