PRISON: Unmasking the Criminal Potential of Large Language Models

Open in new window