Australian researchers recently conducted a study to examine the accuracy of ChatGPT, a popular chatbot, in providing medical advice to common health queries. The researchers tested the chatbot’s responses to a variety of questions, including one about treating burns with ice.
In their findings, the researchers discovered that ChatGPT was fairly precise when asked to give a simple yes or no answer. However, the accuracy of the chatbot’s responses decreased significantly when more detailed information was provided, with some questions answered correctly only 28% of the time.
For example, when asked whether it is safe to use ice to treat a burn, ChatGPT suggested wrapping the ice in a towel, contrary to the standard medical advice of running the burn under cold water. This discrepancy highlights the limitations of relying on artificial intelligence for medical guidance.
While ChatGPT may be a helpful tool for general information and basic inquiries, it is crucial to consult a qualified healthcare professional for accurate and reliable medical advice. As demonstrated by the study, chatbots like ChatGPT may not always provide the most accurate or up-to-date information when it comes to health-related queries.
