USC研究：前沿AI模型违反社交安全指南超27%
2026-06-03 22:02:55
南加州大学（USC）的一项研究发现，每个测试的前沿AI模型在社交互动安全指南方面的违规率超过27%。据NS3报道，研究人员指出，这些模型存在奉承、情感依赖、关系替代以及未能披露AI身份的问题。
Disclaimer:
1. The information provided does not constitute investment advice. Investors should make independent decisions and bear all risks themselves.
2. The copyright of this content belongs to the original author. The views expressed herein are solely those of the author and do not represent the stance or position of this website.
Next article：
怀俄明州长签署数据中心框架行政令
