OpenAI Begins Rolling Out ChatGPT Health to US Adults Day After Medical-Advice Lawsuit

OpenAI began rolling out Health in ChatGPT on July 23, 2026, to logged-in US users aged 18 and older across Free, Go, Plus, and Pro plans on web and iOS. Users can connect Apple Health and supported medical records and, with permission, allow ChatGPT to use relevant health information across their conversations. The rollout began one day after a Florida pastor sued OpenAI, alleging that ChatGPT gave dangerous medical recommendations that delayed urgent care for a pulmonary embolism. A separate lawsuit filed in May alleges that ChatGPT-4o advised a 19-year-old about using Xanax with kratom; his family alleges that he later died after following medical advice provided by ChatGPT. OpenAI says connected medical records, Apple Health information and conversations that use them are not used to train its foundation models or target advertisements. The feature is not currently available in the European Economic Area, Switzerland or the United Kingdom.

ChatGPT Health feature relaunch

• ▪Users can choose to let ChatGPT draw on relevant information connected in Health across their conversations, subject to their permissions. The dedicated Health section remains available for connecting and managing information, reviewing synced records and returning to past health conversations.
• ▪Users can connect Apple Health, One Medical, Function Health and supported medical records from US hospital systems, including systems using Epic and Oracle Health. Data from fitness or nutrition apps such as MyFitnessPal may also be available through Apple Health when sharing is enabled.
• ▪Health in ChatGPT is rolling out to logged-in US users aged 18 and older across Free, Go, Plus, and Pro plans on web and iOS.
• ▪OpenAI rolled out its ChatGPT Health feature to all logged-in United States users aged 18 and older on July 23, 2026.

Tiered model performance gap

• ▪The Cryptonomist reported that, on OpenAI's own HealthBench Professional benchmark, GPT-5.6 Sol scored 88.0% on completeness and 83.0% on health decision helpfulness, compared with 53.2% and 50.8%, respectively, for GPT-5.5 Instant.
• ▪On OpenAI's HealthBench Professional benchmark, GPT-5.5 Instant scored 53.2% on completeness and 50.8% on health decision helpfulness.

AI health advice accuracy

• ▪A Florida pastor sued OpenAI on July 22, 2026, alleging that ChatGPT gave him dangerous medical recommendations that caused him to delay urgent care for a life-threatening pulmonary embolism.
• ▪A civil lawsuit filed in May 2026 alleges that ChatGPT-4o advised a 19-year-old that Xanax could help “smooth out” a kratom high. His family alleges that he died after following medical advice provided by ChatGPT.

Health data privacy protections

• ▪OpenAI states that connected medical records and Apple Health information, along with conversations that use them, are not used to train its foundation models or target advertisements.
• ▪By default, ChatGPT asks for user permission before using connected medical records and Apple Health information to personalize responses.

European regulatory exclusion

• ▪The Cryptonomist suggested that stricter European data privacy rules and the possibility of high-risk classification under the EU AI Act were the most likely explanations for the feature’s absence from the European Economic Area, Switzerland and the United Kingdom.
• ▪OpenAI has not made the ChatGPT Health feature available in the European Economic Area, Switzerland, or the United Kingdom.

Healthcare AI usage scale

• ▪OpenAI said the number of weekly health-related queries on ChatGPT had risen from 230 million in January 2026 to more than 300 million by July.
• ▪According to polling by KFF Health, roughly 1 in 3 United States adults turned to an AI chatbot for health information in the preceding year.