According to MacRumors, OpenAI has officially launched ChatGPT Health, a dedicated and separate section within ChatGPT for health-related questions. The feature allows users to connect various health data services, including Apple Health, Function, MyFitnessPal, and Peloton, for personalized responses. It can also integrate with your medical records to analyze lab results and history, though OpenAI stresses it is not for diagnosis or a substitute for a doctor. The service uses multiple layers of encryption and operates as a separate space, with data not used for model training by default. It’s launching with a waitlist for beta users on Free, Go, Plus, and Pro plans outside the European Economic Area, Switzerland, and the UK, with full rollout on web and iOS in coming weeks. Medical record integration is currently only available in the United States.
How it works and why it matters
So, here’s the basic idea. You go into this new “Health” space in ChatGPT, and you can ask it anything from “help me understand my cholesterol panel” to “suggest a workout based on my recent activity.” If you connect Apple Health, it can see your steps, sleep, and workouts. Connect a service like MyFitnessPal, and it knows what you’re eating. It’s basically trying to be a super-informed, AI-powered health buddy that has context.
The privacy tightrope
Now, this is the biggest hurdle, right? Handing your Apple Health data and medical records to an AI company feels… intense. OpenAI knows this. They’re shouting from the rooftops about enhanced privacy, separate data storage, and default opt-out from model training. And look, that’s good. It’s the bare minimum they had to do to even attempt this. But the real test will be in the security audits and, frankly, whether users trust a company with a famously complex corporate structure with their most sensitive data. It’s a massive bet on their ability to be a steward.
Not a doctor, but a research assistant
OpenAI is being very careful with the disclaimers. This is not for diagnosis. Don’t use it instead of seeing a professional. That’s smart, legally. But the potential value is real. How many people get a PDF of lab results and just stare at the numbers? Having an AI explain what “HDL” and “LDL” mean in plain English, in the context of your own history, could be genuinely useful. It could help you formulate better questions for your actual doctor. That’s the real pitch here: a pre-appointment prep tool.
The rollout and the hurdles
The limited beta and geographic restrictions tell a story. This is a slow, cautious launch for a product that could blow up in their faces if there’s a single major data incident. The European exclusion is particularly telling—those GDPR rules are no joke, especially for health data. And honestly, the medical record integration in the US alone is a huge technical and regulatory mountain to climb. Can they actually parse the messy, non-standardized data from hundreds of different health record systems? That’s a huge “we’ll see.” For now, if you’re curious, you can join the waitlist and read the official announcement. But remember, it’s still an AI. And your health isn’t something you want to get wrong.
