New method aims to keep kids safe from illegal AI-generated content
中文摘要
研究人员开发出新型审计技术，无需诱导非法输出即可检测AI恶意功能，旨在保护儿童免受有害生成内容的侵害。
English Summary
Researchers developed a new auditing technique to test AI models for malicious capabilities without prompting illegal outputs, aiming to protect children from harmful generated content.
Original Excerpt
Researchers developed an auditing technique to test generative AI models for malicious capabilities, without prompting them for illegal outputs.