OpenAI says it plans to expand ChatGPT Health to adult users in the United States across its consumer subscription tiers. The feature is designed to answer health-related questions and, when users choose, draw on information connected from health and fitness services.
The company’s availability claims, rollout details and usage figures have not been independently verified. Even so, the proposed expansion shows how OpenAI wants health features to fit into ordinary ChatGPT conversations instead of remaining confined to a separate part of the service.
Health information could extend beyond a dedicated hub
ChatGPT Health began as a dedicated area where users could ask health questions and connect personal information. OpenAI says the integrations offered during testing included Apple Health, Function and MyFitnessPal, although the scope and availability of those connections have not been independently verified.
The company also says users can connect medical information from systems associated with Epic and Oracle Health, as well as services including One Medical and Function Health. Those integration details remain company-provided claims rather than independently confirmed capabilities.
OpenAI previously directed health-related conversations toward the dedicated Health hub. Its proposed approach would let users choose whether connected information can inform relevant questions elsewhere in ChatGPT. That could include a general conversation about food in which allergy information or other selected health details provide additional context.
OpenAI says testing showed that 70% of health-related questions were asked outside the dedicated hub. That figure has not been independently verified, but it helps explain the product decision: people do not always separate health concerns from broader questions about meals, exercise or daily routines.
The change also puts more emphasis on user controls and health data privacy. People considering the feature will need clear information about which services are connected, when personal details can be used in a conversation and how those connections can be revoked.
OpenAI points to growing demand for health questions
OpenAI says ChatGPT users were submitting roughly 230 million health-related queries per week during an earlier stage of testing and that the figure later reached 300 million. Neither number has been independently verified, so they should be treated as company estimates rather than a confirmed measure of demand.
The company has also claimed progress in the ability of its models to handle health questions. A specific performance claim involving an unreleased or insufficiently documented model could not be verified and is not being treated as established. Benchmark results can indicate how a model responds under test conditions, but they do not by themselves establish that its answers are appropriate for an individual medical situation.
OpenAI says it works with physicians to improve health-related responses. It also says information submitted through ChatGPT Health is not used to train its models. These are company statements, and the practical details may depend on the service’s applicable settings and terms.
The feature still carries firm medical limits
OpenAI’s stated position is that its services are not intended to diagnose or treat health conditions. The company has also encouraged users to verify health information and make medical decisions with professional guidance. The precise wording and application of those policies have not been independently verified here.
That distinction matters as health information becomes available in general chats. A conversational answer can help someone organize questions, understand terminology or review information they have chosen to connect, but the feature is not presented as a replacement for a qualified medical professional.
OpenAI says the broader rollout is intended for logged-in US adults using Free, Go, Plus or Pro plans on the web and iOS. The rollout schedule and complete plan-by-plan availability have not been independently confirmed, so users may encounter differences in access.
The larger product shift is clear: OpenAI wants health context to follow users into relevant conversations rather than live exclusively inside a specialized hub. Whether that approach proves useful will depend not only on answer quality, but also on transparent controls, careful handling of personal information and consistent reminders about the limits of AI-generated health guidance.