Blog & news
ChatGPT found by study to spread inaccuracies when answering medication questions
Nearly 75% of responses from the generative AI chatbot were incomplete or wrong, say researchers

DR GPT™ perspective
A study finding that most of a chatbot’s medication answers were incomplete or wrong is a healthy reminder. These tools sound confident even when they are off, which is dangerous around drugs and dosing. I explain how to keep a clinician between the model and the patient in my healthcare AI books.
