According to Thurrott.com, OpenAI is launching a new ChatGPT Health experience today. The feature is designed to provide better health and wellness advice by leveraging users’ personal medical records and connected app data. It’s launching initially for a small group of Free, Go, Plus, and Pro users, but specifically excludes those in the European Economic Area, Switzerland, and the UK. OpenAI says hundreds of millions of people already ask health questions on ChatGPT each week. US users can upload records and connect apps like Apple Health, MyFitnessPal, and Peloton. The company emphasizes that health data is encrypted, isolated from other chats, and won’t be used to train its AI models.
Privacy-first or marketing spin?
Look, the privacy promises here are strong on paper. Isolated data, purpose-built encryption, and a pledge not to use conversations for training. That’s the bare minimum you’d expect, frankly, but it’s good they’re stating it upfront. The real test will be in the implementation and, let’s be honest, future policy changes. Can users *really* trust a company whose core business is data-hungry AI with their most sensitive information? The immediate exclusion of Europe and the UK is telling—those regions have GDPR and other strict data laws that make this kind of product a compliance nightmare. So OpenAI is starting where the regulatory waters are, shall we say, more navigable.
What this means for users
Basically, if you’re in the US and get access, you’re being offered a very personalized health chatbot. Connect your Apple Health, upload a PDF of your lab results, and you can ask things like “What do these cholesterol numbers mean for me?” or “Suggest a workout based on my recent heart rate data.” The integration with apps like Instacart for groceries or AllTrails for hikes tries to move it from passive advice to active lifestyle management. Here’s the thing: it’s positioned to “support, not replace” clinicians. But will people use it that way? Or will it become a source of anxiety and armchair diagnosis? The track record of AI in medicine is… mixed. It can be a powerful tool for understanding, but it’s notoriously bad at nuance and can hallucinate just like in any other topic.
The bigger picture: AI in your medical chart
This is a massive strategic land grab by OpenAI. Health data is the ultimate sticky, valuable dataset. By creating a dedicated, walled garden for it, they’re not just building a feature—they’re building a new platform. They’re going straight after the traditional health tech and telemedicine players. Think about it: if ChatGPT Health becomes your primary interface for understanding your health, where does that leave your doctor’s patient portal? Or standalone wellness apps? It centralizes everything under OpenAI’s roof. For now, it’s optional and separate. But the potential for this to evolve into a paid health subscription service, or a data pipeline for future medical AI models (with consent, they’d say), is enormous. This isn’t just a chatbot upgrade. It’s a foothold in an industry worth trillions.
