Explore the persona paradox in clinical LLMs: how assigned roles impact AI performance, safety, and business outcomes. Learn why context-dependent evaluation is crucial for enterprise AI deployment.