跳到正文
Boris Cherny· X:Boris Cherny (@bcherny)·· 3 小时前AI 评分38

Anthropic 与 Gray Swan 合作举办 30 万美元间接提示词注入红队挑战赛

AI 导读

Anthropic 与 Gray Swan 合作举办奖金 30 万美元的间接提示词注入红队挑战赛,面向业界征集针对 Claude 模型的攻击发现。Anthropic 称过去几年在 Claude 的提示词注入防护上已取得显著进展,但随着智能体承担的任务增多,跨行业解决该问题愈发紧迫。

正文

Over the last few years, we have made significant progress on prompt injection for Claude models. As agents do more, it is increasingly urgent that we solve it across the industry.

Really excited to partner with Gray Swan to host this competition. Can’t wait to see what you find!

来源:Boris Cherny · x.com