Microsoft is bringing Claude into healthcare workflows through Foundry — aiming at clinical and administrative tasks that require accuracy and compliance. Does AI in regulated medical settings actually improve reliability, or just shift risk?

I think the actual story here is why Claude and not OpenAI?