A Realistic Threat Model for Large Language Model Jailbreaks

Open in new window