OpenAI Unveils Segregated “ChatGPT Health” After 40 M U.S. Users Turn to Bot Daily for Medical Advice

Every generation gets a new bedside manual: Mayo’s “Book of Popular Science” (1937), the Merck Manual on CD-ROM (1995), WebMD’s symptom checker (1999), and now an always-on large-language-model. Like those earlier leaps, ChatGPT Health rides two long arcs—democratization of medical knowledge and the steady digitization of personal health data—but it also concentrates unprecedented power over private records in one commercial platform. The move echoes how Google’s 2008–2012 Health project tried (and failed) to become patients’ data hub, underscoring that trust, standards and regulation, not just software, decide adoption. If generative AI becomes the default interpreter of lab results and insurance bills, it could, over decades, erode traditional gatekeepers’ control of information and reshape a U.S. healthcare market that already absorbs nearly 18 % of GDP; yet the same centralization heightens 21st-century risks of data breaches, algorithmic bias, and depersonalized care. Measured against a 100-year timeline, this launch may mark the moment health literacy shifted from static references to dynamic, personalized dialogue—provided society can align privacy rules and clinical validation as fast as the technology spreads.