OpenAI News·· 2025-11-07AI 评分52
OpenAI 探讨 AI 系统中的提示注入攻击
Understanding prompt injections: a frontier security challenge
AI 导读
OpenAI 讨论了 AI 系统中的提示注入攻击，并介绍了其在安全研究、模型训练和用户防护方面的进展。
原文
Prompt injections are a frontier security challenge for AI systems. Learn how these attacks work and how OpenAI is advancing research, training models, and building safeguards for users.
来源：OpenAI News · openai.com