ChatGPT Health Opens to All US Users, But Its Own Team Walks Back the 'Better Than Doctors' Claim
OpenAI is bringing its medical chatbot to every logged-in American adult today. One executive said the AI reasons better than clinicians. Another quickly disagreed.
Key points
- OpenAI launched ChatGPT Health to all US users aged 18 and older on Thursday, covering free, Go, Plus, and Pro plans.
- An OpenAI vice president said its models reason "at levels that are better than clinician level"; a second executive immediately softened that claim.
- A Florida pastor sued OpenAI on Wednesday, alleging ChatGPT gave him dangerous medical advice that led him to delay treatment for a pulmonary embolism, a blood clot in the lungs.
- ChatGPT Health can connect to Apple Health, Weight Watchers, MyFitnessPal, lab results, and post-visit notes, with additional encryption on stored health data.
- OpenAI states the tool "supports, not replaces, professional care."
OpenAI's health chatbot is now available to every adult in the United States with a ChatGPT account, and the rollout arrived with a bold claim that the company quickly had to walk back.
At a press briefing, Ashley Alexander, OpenAI's vice president of health product, said the company's models "are now capable of reasoning at levels that are better than clinician level." That is a striking sentence. Within the same briefing, a colleague stepped in.
Karan Singhal, OpenAI's health lead, said he would "temper" that claim. He pointed to individual studies, including one from Harvard and Stanford, that point in that direction but stopped well short of declaring the AI better than your doctor.
What does ChatGPT Health actually do?
It connects your personal medical records and health-tracking data to a chatbot so you can ask questions about your own health in plain English. Until now, that feature lived in a separate tab. With Thursday's update, it sits inside the main chat window, making it much easier to reach.
The types of data you can connect include post-visit notes, messages from your care team, lab results, and your current medications. Supported services include Apple Health, Weight Watchers, MyFitnessPal, and Function, along with records from health providers. OpenAI says ChatGPT will ask your permission before pulling health data into any given conversation, and that all conversations are encrypted, with health records receiving an extra layer of protection on top.
The feature runs on OpenAI's newest models, including GPT-4o Sol, which the company describes as its strongest model yet for health.
Should users be worried?
That depends on how you use it. OpenAI is plain about the limit: this tool supports professional care, it does not replace it. Think of it as a very well-read assistant that can help you understand a lab result or prepare questions before an appointment, not a replacement for calling your GP.
The timing of the launch makes that boundary feel especially important. The day before the rollout, first reported by The Verge AI, a Florida pastor filed a lawsuit against OpenAI. He claims ChatGPT gave him "extremely dangerous medical recommendations" and that he delayed getting treatment for a pulmonary embolism, a potentially fatal blood clot in the lungs, as a result.
OpenAI has not publicly commented on the lawsuit.
Common questions
Who can use ChatGPT Health today?
Any logged-in ChatGPT user aged 18 or older in the United States, on free, Go, Plus, or Pro plans, on the web or on iOS.
Is my health data safe?
OpenAI says health data receives additional encryption beyond its standard protections, and the chatbot asks permission before using your records in a conversation. As with any digital health tool, check what you are consenting to share before connecting accounts.
Can ChatGPT replace a real doctor?
No. OpenAI states explicitly that the product supports professional care rather than replacing it. Use it to understand information or prepare for appointments, then speak to a qualified clinician for any diagnosis or treatment decision.
