1
PRISON: Unmasking the Criminal Potential of Large Language Models
一项新框架系统评估大语言模型在真实互动中的犯罪潜力，直指AI安全暗面。
arXiv:2506.16150v4 Announce Type: replace-cross Abstract: As large language models (LLMs) advance, concerns about their misconduct in complex social c…