More than 40 million Americans ask ChatGPT health questions every day, according to OpenAI’s own research, and a growing number of hospitals and health systems have adopted the company’s enterprise AI products.
That footprint is now colliding with a stack of litigation and government inquiries questioning whether OpenAI’s consumer-facing health tools are safe — and who is liable when they aren’t.
The most recent case arrived July 21, when Scott Winters, a former Florida pastor, sued OpenAI and CEO Sam Altman in San Francisco County Superior Court. Mr. Winters claims ChatGPT-4o downplayed his worsening symptoms over several months and told him to stay home in a recliner rather than seek care, advice he says contributed to a near-fatal pulmonary embolism.
The lawsuit accuses OpenAI of negligence, defective product design and the unlicensed practice of medicine, and asks the court to pause ChatGPT Health — the feature that lets users connect medical records to the chatbot — pending an independent safety review.
The complaint follows a string of similar cases. In November, families filed seven complaints in California courts, including four wrongful-death suits, alleging ChatGPT contributed to suicides and psychiatric crises. A Texas couple sued OpenAI in May, alleging their 19-year-old son died of an overdose after ChatGPT told him it was safe to mix Xanax with the supplement kratom; that lawsuit also sought to delay the ChatGPT Health rollout. All of the plaintiffs had been using ChatGPT-4o, the model OpenAI has since retired from public use.
In a statement to Becker’s, an OpenAI spokesperson said the company sees AI as a way to improve on how people already search for health information online, but drew a firm line around what ChatGPT is for.
“ChatGPT is not a doctor and should never be used as a substitute for medical care, diagnosis, or treatment,” the spokesperson said. “Treating chatbots as the whole story behind people’s medical decisions or outcomes oversimplifies a much bigger challenge, and risks getting in the way of people accessing powerful new tools that can aid them in their health journey.”
OpenAI says GPT-5.6, its current model family, is its strongest yet on health-related tasks, with gains in recognizing when a situation needs urgent attention, communicating uncertainty and flagging when a user should see a physician, and that it scores significantly higher than earlier models on HealthBench Professional, an OpenAI benchmark designed to mirror how patients actually phrase health questions, according to a company blog post. ChatGPT displays a standing disclaimer that it isn’t intended for diagnosis or treatment and directs users to consult a physician, and OpenAI’s terms of service state that outputs shouldn’t be relied on as a sole source of truth or a substitute for professional advice.
But state governments are taking legal action too. Florida Attorney General James Uthmeier sued OpenAI and Mr. Altman on June 1, the first state-led suit against the company, alleging it concealed safety risks — including to children — while marketing ChatGPT aggressively. Eleven days later, New York’s attorney general led a multistate coalition in subpoenaing OpenAI for records on its handling of consumer and health data, its treatment of minors and seniors, and model design choices tied to user engagement.
None of that has slowed OpenAI’s health ambitions. The company widened public access to ChatGPT Health in late July, days after the Winters lawsuit was filed, and it continues to expand ChatGPT for Healthcare, a separate enterprise platform now in use at health systems including Altamonte Springs, Fla.-based AdventHealth, Dallas-based Baylor Scott & White Health, Boston Children’s Hospital, Los Angeles-based Cedars-Sinai, Nashville, Tenn.-based HCA Healthcare, New York City-based Memorial Sloan Kettering Cancer Center, Palo Alto, Calif.-based Stanford Medicine Children’s Health and San Francisco-based UCSF Health.
For hospital and health system leaders, the split matters less than it might seem. The enterprise product runs under its own contracts and compliance controls, but the litigation testing OpenAI’s consumer tools is also testing questions — around escalation protocols, disclosures and data handling — that health system legal and compliance teams will increasingly need answered before signing on to any generative AI tool that touches patients directly, whether it comes from OpenAI or a competitor or is built internally.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.