TL;DR

OpenAI introduced ChatGPT Health, a standalone experience that lets users securely connect medical records and wellness apps to ChatGPT while adding layered privacy protections. The product was developed with input from hundreds of physicians and is intended to support — not replace — clinical care.

What happened

OpenAI announced ChatGPT Health, a separate area within ChatGPT intended to combine users' personal health information with the assistant's capabilities. The experience supports secure connections to medical records and popular wellness apps so responses can reflect a user’s own data. Health operates with added protections: conversations are isolated in a distinct space with separate memories, enhanced encryption and purpose-built isolation, and Health chats are not used to train foundation models. The feature was created with clinical input from more than 260 physicians across 60 countries, who reviewed model outputs extensively. Early access will be limited to a small group of users; eligibility initially includes Free, Go, Plus, and Pro plan holders outside the European Economic Area, Switzerland, and the United Kingdom. Some integrations and app connections are available only in the U.S., and syncing Apple Health requires iOS.

Why it matters

  • Brings personal health data into conversational AI, which can make explanations and appointment prep more context-aware.
  • Adds layered privacy and isolation specifically for sensitive health conversations, including a promise not to use Health chats to train foundation models.
  • Incorporates sustained physician oversight and a clinical evaluation framework to guide safety, clarity, and appropriate escalation.
  • Could change how people gather and organize dispersed health information from portals, wearables, and apps ahead of clinical visits.

Key facts

  • Over 230 million people worldwide ask health and wellness questions on ChatGPT each week (de-identified analysis cited).
  • ChatGPT Health stores conversations, connected apps, and files separately from other chats and uses separate memories for health context.
  • Conversations and files across ChatGPT are encrypted at rest and in transit; Health adds purpose-built encryption and isolation for sensitive data.
  • Health conversations are not used to train OpenAI’s foundation models.
  • OpenAI collaborated with more than 260 physicians from 60 countries; that network provided feedback on outputs over 600,000 times across 30 focus areas.
  • OpenAI partners with b.well to enable access to trusted U.S. healthcare providers and medical records integrations.
  • Apps named for Health connections include Apple Health (iOS only), Function, MyFitnessPal, Weight Watchers, AllTrails, Instacart, and Peloton.
  • Early access is being offered to a limited group; eligible users initially are Free, Go, Plus, and Pro subscribers outside the EEA, Switzerland, and the UK.
  • Users can remove connected apps and delete Health memories; enabling multi-factor authentication is recommended to strengthen account security.

What to watch next

  • Planned expansion to all users on web and iOS in the coming weeks, as OpenAI refines the experience with early users.
  • Whether medical-record integrations and additional apps will become available outside the U.S. (not confirmed in the source).
  • How ongoing physician feedback and the HealthBench evaluation framework affect updates to model behaviour and safety checks.

Quick glossary

  • Multi-factor authentication (MFA): A security measure that requires two or more verification methods (for example password plus a code) to access an account.
  • Encryption at rest and in transit: Techniques that protect data both when it is stored on servers (at rest) and when it is sent over networks (in transit).
  • Medical records integration: A connection that allows an app or service to access a user’s clinical data, such as lab results, visit summaries, and medical history, with user permission.
  • Foundation models: Large-scale AI models trained on broad data that can be adapted for different tasks, such as conversational assistants.
  • HealthBench: An evaluation framework mentioned by OpenAI that assesses model outputs against physician-written rubrics reflecting clinical priorities like safety and clarity.

Reader FAQ

Is ChatGPT Health intended to replace doctors?
No. OpenAI says Health is designed to support users and help them prepare for clinical conversations, not to provide diagnosis or replace clinicians.

Will conversations in ChatGPT Health be used to train OpenAI’s models?
No. The announcement states Health conversations are not used to train foundation models.

Who can get access to ChatGPT Health now?
OpenAI is starting with a small group of early users; eligibility initially includes Free, Go, Plus, and Pro plan holders outside the EEA, Switzerland, and the UK.

Can I disconnect apps or remove my health data?
Yes. Users can remove app access at any time and view or delete Health memories as described in Settings.

Will ChatGPT Health be available in the European Economic Area, Switzerland, and the UK soon?
Not confirmed in the source.

January 7, 2026 Product Introducing ChatGPT Health A dedicated experience in ChatGPT designed for health and wellness. Join the waitlist (opens in a new window) 00:0002:31 Listen to article 10:16…

Sources

Related posts

By

Leave a Reply

Your email address will not be published. Required fields are marked *