Designing AI agents to resist prompt injection

媒体报道 OpenAI 原文语言:英文

原文摘录

How ChatGPT defends against prompt injection and social engineering by constraining risky actions and protecting sensitive data in agent workflows.

以下是来源提供的简短摘录,完整内容请查看原文。

参考来源

← 返回每日AI报告