What to Know Before Asking an AI Chatbot for Health Advice

March 18, 2026

Technology companies are now offering specialized health-focused chatbots, with OpenAI launching ChatGPT Health in January and Anthropic providing similar features through Claude. These AI programs can analyze medical records, wellness data, and wearable device information to help users understand test results and prepare for doctor visits, though companies emphasize they shouldn't replace professional medical care. While some medical experts view these tools as improvements over generic internet searches due to their personalized responses, significant concerns remain about privacy protections since chatbot companies aren't covered by HIPAA regulations that govern healthcare providers.

Who is affected

Hundreds of millions of people using chatbots for health advice
Users of ChatGPT Health (currently on a waiting list)
Some users of Anthropic's Claude chatbot
Patients seeking health information online
Doctors and healthcare providers
AI companies including OpenAI, Anthropic, and Google
Healthcare researchers studying AI effectiveness

What action is being taken

OpenAI is offering ChatGPT Health with a waiting list system
Anthropic is providing similar health features for some Claude users
Both OpenAI and Anthropic are keeping users' health information separate from other data types and not using it to train their models
Doctors and researchers are testing and evaluating these health chatbot programs
Some medical professionals, like Dr. Wachter, are consulting multiple chatbots to verify information

Why it matters

This represents a significant shift in how people access health information, offering more personalized guidance than traditional internet searches by incorporating individual medical history and context. However, the privacy implications are substantial since chatbot companies operate outside HIPAA protections that govern traditional healthcare providers, meaning users' sensitive medical data lacks the same legal safeguards. The technology's mixed performance in studies—excelling at written scenarios but struggling with real human interactions—highlights both its potential and current limitations for making sound health decisions.

What's next

No explicit next steps stated in the article

Read full article from source: The San Diego Voice & Viewpoint