A study conducted by researchers in the UK has found that artificial intelligence chatbots are often mistaken for health information, often guiding users to correct diagnoses or decisions. The study, which included over 1,200 participants with no medical training, found that chatbots sometimes presented false information or changed its advice depending on the user's wording. The researchers tested commercially available chatbots like OpenAI’s ChatGPT and Meta’�s Llama. However, they found that these chatbots were no better than the control group. The experiment was not a perfect window into how chatbots answer medical questions in the real world. The lead author of the paper suggested that chatbot should ask follow-up questions, similar to the way doctors gather information from patients.