ChatGPT's Accuracy in Answering Medication Questions Questioned by Study
A study conducted by pharmacists found that OpenAI's ChatGPT provided incomplete or inaccurate answers to nearly three-fourths of medication-related questions. The study compared ChatGPT's responses to those provided by pharmacists and found that out of 39 questions, only 10 received satisfactory answers. In some cases, ChatGPT provided incorrect information about drug interactions, highlighting the potential risks of relying on AI for medication-related information. The study emphasizes the importance of verifying information from trusted sources and urges healthcare professionals to evaluate the appropriateness and validity of AI tools in medication-related use cases. OpenAI advises users not to rely on ChatGPT's responses as a substitute for professional medical advice.