OpenAI’s ChatGPT Health requests medical files and cybersecurity pros push back

A chatbot that knows your medical history

Imagine asking a health question and getting an answer tailored to your unique medical past. OpenAI launched ChatGPT Health, a personalized AI assistant that can connect to users’ medical records and wellness apps to provide tailored explanations and summaries.

It’s designed to support your wellness journey by understanding your specific data. This move is sparking major conversations about the future of healthcare accessibility and digital privacy.

Upload your records for personalized answers

The feature lets eligible users connect medical records and supported wellness apps, such as Apple Health, and can summarize bloodwork, explain common lab results, and highlight trends over time for easier discussion with a clinician.

The goal is to help you better understand your information before a doctor’s appointment. OpenAI clearly states it is not for diagnosis or replacing professional care.

Your HIPAA protection may vanish

Sharing records with this AI likely removes their protection under HIPAA, the strict health privacy law. Your data’s security then depends on the company’s own policies, not federal regulation.

Privacy experts urge caution, as these corporate promises can change. Sensitive information could be used in unexpected ways in the future.

Why millions are asking AI for health help

The appeal lies in immediate, accessible, and free answers without waiting rooms or judgment. People often seek explanations for confusing medical terminology or to understand basic symptoms.

For people with limited access to care, a conversational tool can feel like an immediately available resource; OpenAI says tens of millions of users ask health-related questions on the platform each month.

The risk of AI hallucinations in health

These confident fabrications are especially perilous in medical contexts where accuracy is critical. The AI might generate plausible-sounding but incorrect explanations for lab values or symptoms.

Its design to be helpful and provide answers can suppress crucial I don’t know responses. This inherent unreliability means any advice must be scrutinized and confirmed by a qualified professional.

How real doctors helped shape the tool

OpenAI says it consulted hundreds of clinicians, and clinicians provided feedback to improve clarity, urgency triage language, and flag situations that require immediate medical attention, but this consultation is not the same as formal clinical validation.

This medical guidance aims to steer the tool toward safe, educational support rather than speculative diagnosis. Their involvement is a key part of the product’s claim to be a responsible ally.

Your data could become a corporate asset

In a bankruptcy or acquisition, user data is often considered a transferable asset. This precedent means your intimate health details could be sold to another company you didn’t initially choose.

The long-term fate of your digital health footprint is tied to the company’s financial health, not your own. It introduces a commercial dimension to the stewardship of your most personal information.

A global rollout with notable exclusions

Regions like the EU with strong GDPR protections are excluded, signaling the legal complexity of processing health data. This creates a geographic disparity in who can access such tools and under what safeguards.

It also pressures other regions to develop clear regulations for health AI. The rollout map clearly reflects the world’s uneven landscape of digital privacy rights.

The race for your AI health companion is on

Tech giants are investing heavily, seeing health as a key domain for AI applications and user engagement. This competition fuels rapid feature development but also a race to capture valuable user data.

Different companies will market varying safety and privacy promises to win your trust. Consumers will need to critically compare these claims as the market grows more crowded.

Using it wisely, a guide for patients

Think of it as a tool for organizing information and formulating questions, not for making decisions. It is well-suited for explaining general medical concepts or clarifying notes from a past visit.

Steer clear of using it for interpreting new, acute symptoms or adjusting medication. Establishing clear personal boundaries for its use is essential for safety.

Regulators are taking a hands off approach

At present, many regulators have allowed consumer health tools to develop subject to consumer protection rules and existing law, while some experts and policymakers are calling for stronger targeted oversight of health-related AI.

The regulatory environment could change swiftly if high-profile harms occur. For now, navigating this space requires a healthy dose of personal caution and research.

The final verdict on AI health assistants

They excel as administrative and educational aids, helping users manage information overload. These assistants can be useful administrative and educational tools, but they do not replace clinical judgment, and users should confirm any medical interpretation with a qualified clinician.

The value they provide must be constantly weighed against privacy compromises and accuracy risks. They are supplemental instruments, not foundational elements, of good healthcare.

Want to see how these AI slip-ups play out in real life? Check out how ChatGPT’s diagnosis errors challenge doctors on the frontlines.

Your health, your informed choice

Understand the trade-off, convenience, and insight in exchange for potential data vulnerability. Begin by using it with non-sensitive information to gauge its utility for your life.

Regularly review the privacy settings and terms of service, as they may evolve. Empower yourself with knowledge, and let that guide your interaction with this powerful, yet imperfect, technology.

Curious about simple steps to lock things down? See the ChatGPT privacy tips most users overlook but should enable today.

What’s your take on using AI for health guidance? Share your thoughts in the comments, and if you found this helpful, give it a like.

This slideshow was made with AI assistance and human editing.

Don’t forget to follow us for more exclusive content right here on MSN.

Read More From This Brand:

• Sam Altman flags privacy risks in ChatGPT therapy
• Free Chinese AI model beats GPT-5 and Sonnet 4.5
• OpenAI introduces quirky ChatGPT personality to spice up conversations