OpenAI expands ChatGPT Health to US users amid safety claims and legal scrutiny
The company asserts its AI reasoning exceeds clinician levels, though health lead Karan Singhal tempers the claim with academic references. The rollout follows a lawsuit alleging dangerous medical advice.
OpenAI has launched ChatGPT Health to all logged-in users aged 18 and older in the United States, integrating the feature directly into the main chat experience across web and iOS platforms. The update is available across all subscription tiers, including free, Go, Plus, and Pro plans, and is powered by the company’s GPT-5.6 Sol model, which OpenAI describes as its strongest model for health applications to date.
The new functionality allows users to connect medical records and health-tracking data from various providers and third-party services, including Apple Health, Weight Watchers, MyFitnessPal, and Function. Unlike previous iterations that required navigating to a dedicated tab, the feature now operates within the primary chat interface. Users must grant explicit permission before their health information is used to personalise responses, and they retain the ability to review and modify the specific types of data connected, such as post-visit notes, care team notes, lab results, and medications.
During a briefing, Ashley Alexander, OpenAI’s vice president of health product, stated that the company’s models are now capable of reasoning at levels superior to clinicians. However, OpenAI health lead Karan Singhal tempered this assertion during the same session, noting that while individual studies from Harvard and Stanford point in that direction, he would not definitively assert superior reasoning capabilities without further qualification.
The broad release comes shortly after a lawsuit was filed by a Florida pastor against OpenAI. The legal action alleges that ChatGPT provided extremely dangerous medical recommendations that led the plaintiff to delay care for a pulmonary embolism. In response to growing scrutiny, OpenAI has implemented additional encryption protections for health data connected through the Health feature, going beyond the standard encryption at rest and in transit applied to all conversations.
OpenAI maintains that ChatGPT Health is designed to support, not replace, professional medical care. The service remains restricted to users in the US aged 18 and older, with the company emphasising that the AI acts as an adjunct to human expertise rather than a substitute for it.
