Why AI Isn't Enough
Don't take our word for it.
We asked ChatGPT directly about its own Health product. Here's what it said.
ChatGPT's response:
"The honest bottom line"
"If these were pitched honestly, it would be:"
"We are building a health product that happens to use AI."
"We are building AI that happens to help with health."
— ChatGPT's own analysis, January 2026
On legal protection:
"HIPAA gives you stronger procedural armor."
"The legal shield is objectively thinner [for ChatGPT Health]."
"Yes. The legal risk of disclosure is higher for consumer AI conversations than for HIPAA-protected PHI."
On subpoenas and legal exposure:
"If someone is risk-minimizing during custody litigation, it is rational to prefer HIPAA-protected channels for sensitive mental health matters."
"Courts do treat PHI differently than consumer data."
"HIPAA doesn't just add friction — it changes the burden."
On choosing where to start:
"There is no rational privacy argument for starting with ChatGPT Health for health matters that may matter later."
"Starting with a HIPAA platform is the correct move."
"And you lose nothing functionally."
ChatGPT's own description of its role:
"ChatGPT Health remains useful as:
- a secondary explainer
- a scratchpad
- a low-stakes thinking aid
—but not as a primary health reasoning system once parity exists elsewhere."
ChatGPT Health User Stories
January 26, 2026 | Geoffrey A. Fowler, Ryan Christoffel
"Apple Watch user gave ChatGPT Health his data, with troubling results"
"It gave me an F. I freaked out and went for a run. Then I sent ChatGPT's report to my actual doctor. Am I an F? 'No,' my doctor said. In fact, I'm at such low risk for a heart attack that my insurance probably wouldn't even pay for an extra cardio fitness test."
"When I tried asking the same heart longevity-grade question again, suddenly my score went up to a C. I asked again and again, watching the score swing between an F and a B."
"Across conversations, ChatGPT kept forgetting important information about me, including my gender, age and some recent vital signs."Read full article →
January 27, 2026
"Do not connect ChatGPT to Apple Health. Not now when you have to apply to test it out, and not when it's rolled out publicly like a finished and working tool."
AppleInsider tested ChatGPT with 10 years of health data. ChatGPT gave an F grade. The user's doctor said he was so low-risk his insurance wouldn't even pay for a test. When asked the same question multiple times, ChatGPT gave every grade from F to B.
Read full article →University of Sydney Researchers
"Independent research consistently shows generative AI tools do sometimes give unsafe health advice, even when they have access to a medical record."
Academic researchers from the University of Sydney analyzed ChatGPT Health and found significant concerns about accuracy, privacy, and safety. The researchers note that "AI responses can often seem sensible – and increasingly personalised – but that doesn't necessarily mean they are correct or safe."
They call for purpose-built AI health tools that "prioritise accuracy, equity and transparency" — exactly what Hale was designed to provide.
Read full research →January 30, 2026
"Here's What Doctors Should Know About ChatGPT Health"
OpenAI spokesperson confirms to medical publication MedPage Today that "HIPAA doesn't apply to ChatGPT Health." The article reports that a Washington Post reporter received an F-grade in cardiac health from ChatGPT Health, but his real doctor said "this couldn't be further from the truth."
Dr. Henry Bair of Wills Eye Hospital warns that ChatGPT Health "can derail a lot of visits, because you have to spend a lot of time on parsing out what is valid and what's not."
Dr. Gregory Marcus of UCSF notes that "the reliability and accuracy of any inferences still relies on the veracity of the information used to train these algorithms."Read article →
The comparison
| Hale | ChatGPT | |
|---|---|---|
| HIPAA Compliant | ✓ | ✗ |
| Signed BAAs | ✓ | ✗ |
| PHI Legally Protected | ✓ | ✗ |
| Zero Data Retention | ✓ | "Not currently" |
| Data Used for AI Training | No | "Not currently" |
| Built for Healthcare | ✓ | ✗ |
| Knows Medical HistoryKnows Your Medical History | ✓ | ✗ |
Note: Because your health information in ChatGPT is not legally classified as PHI, they can change their retention and training policies at any time without notice.
What ChatGPT admitted:
Your health data has weaker legal protection
Consumer AI conversations don't have the statutory protections that HIPAA-covered PHI has.
Your data can be subpoenaed
In custody disputes, employment cases, or insurance investigations, your ChatGPT health conversations are discoverable with standard legal process.
HIPAA platforms are the better choice
For anyone who cares about legal protection, portability, or future exposure, "starting with a HIPAA platform is the correct move."
You lose nothing by choosing Hale
Same capabilities, stronger protection.
Their advice on using ChatGPT for health:
"Don't make ChatGPT your long-term mental health journal."
We built Hale so you don't have to worry about that.
Read the full conversations
Don't just take our word for it — see the raw ChatGPT conversations yourself.
Ready to protect your health data?
No credit card required. HIPAA compliant from day one.
About This Conversation
We asked ChatGPT directly about HIPAA protection, legal exposure, and when to use ChatGPT Health vs HIPAA-compliant platforms. This is an edited transcript of that conversation, condensed for clarity and focus. We removed repetitive sections and some back-and-forth, but all quotes are accurate and in context.
Date: January 23, 2026
Full conversation available on request.