What are the primary risks associated with deploying generative AI models in a healthcare setting, and how do you mitigate them?
