According to CNET, a MacRumors reporter named Adam Perris spotted the Apple Health icon hidden within the ChatGPT app code on Monday, May 27th. The icon was reportedly labeled with images related to activity, sleep, diet, breathing, and hearing. This suggests the Apple Health app will soon be able to connect to OpenAI’s ChatGPT. If this happens, the AI chatbot could access your health and fitness data to provide more customized answers to health questions. It remains unconfirmed when or if this integration will launch, and representatives for both Apple and OpenAI did not immediately respond to requests for comment. Currently, ChatGPT is already integrated with several third-party programs like Google Drive, Slack, and Dropbox through a feature called “apps.”
Privacy and Hallucination Soup
So, let’s just state the obvious here. This is a phenomenally bad idea waiting to happen. The report notes that security and privacy safeguards are “unclear.” That’s putting it mildly. Apple Health contains some of our most sensitive personal data. The idea of piping that into a cloud-based AI model, even with consent, should make anyone’s privacy spidey-sense tingle. Sure, Apple is famously strict about data, but OpenAI’s track record is, well, different. The question isn’t just about encryption during transfer. It’s about what happens to that data once it’s in OpenAI’s systems for processing. Do you really want your sleep patterns or heart rate variability becoming part of an AI training corpus? I don’t.
Experts Are Screaming Into the Void
And here’s the thing: even if we solve the privacy nightmare, we’re left with the accuracy catastrophe. The article points out that experts are “increasingly raising alarms” about using chatbots for health. That’s a polite way of saying doctors and therapists are horrified. ChatGPT is not a medical professional. It’s a statistical pattern machine that is notoriously prone to confident hallucinations. Even OpenAI execs tell you not to trust everything it says. Imagine someone with anxiety symptoms asking for advice and getting a plausible-sounding but completely fabricated list of terrifying rare diseases. Or someone skipping a real doctor’s visit because ChatGPT gave them generic, incorrect diet advice based on their Apple Health data. The potential for harm is massive and real.
Why This Exists At All
Now, I get why they’re exploring it. On paper, it sounds powerful. “Hey ChatGPT, analyze my sleep, activity, and nutrition data from last month and suggest improvements.” That’s a compelling wellness pitch. But that’s exactly the problem—it dresses up a dangerously unreliable tool in the clothing of personalized care. It blurs the line between a fun tech demo and a health advisory service. Basically, it gives a veneer of authority to a system that doesn’t deserve it. We’re already bad enough at seeking proper medical help. Integrating health data with a chatbot that can lie with a straight face feels like building a highway to misinformation. They might launch it with a thousand disclaimers, but you know people will ignore them. They just will.
