What to know before asking an AI chatbot for health advice

With the surge in AI chatbot use for health inquiries, notably OpenAI’s launch of ChatGPT Health designed to analyze medical records and wellness data, experts emphasize caution when relying on these technologies for medical advice. AI chatbots show promise in quickly providing accessible health information, yet several experts and studies highlight significant limitations and risks. For example, Duke University’s Monica Agrawal points out that many AI models are tested on idealized exam questions, but real patient inquiries are more complex, emotional, and nuanced, sometimes leading to misleading or risky advice. A recent Oxford University study found that people who used AI chatbots for hypothetical health scenarios did not make better health decisions than those using traditional online searches or personal judgment.

Doctors stress that AI chatbots lack the ability to fully understand an individual’s complete medical history, which is vital for accurate diagnosis and safe medical guidance. They advise users to share as many relevant health details as possible with AI tools to improve accuracy but warn against relying solely on AI for significant health decisions. Instead, AI should be viewed as a tool to supplement, not replace, professional medical consultation. This caution reflects ongoing concerns about AI chatbots providing inconsistent or inaccurate advice, as their output can vary depending on how questions are phrased.

Overall, medical professionals recommend verifying AI-generated health information with qualified healthcare providers to ensure safe and personalized care.