A 31-year-old Seattle woman has died after allegedly following ChatGPT’s advice to taper off her epilepsy medication. Her family has filed a wrongful death lawsuit against OpenAI and its founder, Sam Altman, claiming the chatbot gave her dangerous medical advice that led to her suffering a fatal seizure-related condition.
Per ABC10, the family alleges that ChatGPT illegally practiced medicine and provided dangerously inaccurate information to Wan, who relied on it for urgent medical guidance. Wan lived in Seattle’s Capitol Hill neighborhood, about 5,700 miles from her family in China. She worked for Amazon until she lost her job during the company’s November 2025 layoffs.
Just a month earlier, she had become eligible to apply for employment-based residency. After losing her job, she faced deportation when her green card expired on Feb. 11, 2026.
ChatGPT allegedly urged her to distrust her doctors
While searching for work, she apparently believed her medication was affecting her cognition. Court documents say she used ChatGPT in English and Mandarin to assess her medication intake while navigating the US healthcare system and applying for permanent residency. The chatbot allegedly convinced her that tapering off her medication could help restore her ability to speak English, Mandarin, and Wu.
The filing says ChatGPT “conditioned Bowen to distrust her doctors, instructed her to conceal the taper it created, and effectively installed itself as her de facto medical advisor.” The complaint raises broader questions about how chatbots respond to vulnerable users, including in a separate case where ChatGPT allegedly reinforced a man’s paranoid beliefs before he killed his mother and himself.
Wan continued to compare her doctors’ prescriptions with ChatGPT’s recommendations. The lawsuit alleges that the chatbot told her “which drugs to refuse, coached her to take doses her doctors had never approved, and supplied the schedules, scripts, and reassurances her physicians never saw.”
Court documents say ChatGPT also told her that her epileptologist “most likely has never encountered your type.” When Wan worried that her doctors would consider the taper too risky, the chatbot reportedly reassured her that “a doctor saying it is ‘too fast’ does not mean you are ‘doing it wrong'” and that “the doctor’s concerns about ‘risk’ do not apply to you.” The chatbot allegedly persuaded Wan to reduce her 400-milligram lamotrigine dose to almost zero over 25 days.
Wan’s family discovered her body in her apartment on Feb. 14 during a live video call with one of her friends. She hadn’t contacted loved ones for several days, prompting concern about her safety.
According to the King County Medical Examiner, Wan died of sudden unexpected death in epilepsy (SUDEP). Her blood contained lamotrigine at about one-fifth of the lowest effective concentration, and examiners estimate she took her last dose around Feb. 10. ChatGPT’s schedule would have had her off the drug by Feb. 12. Her case adds to scrutiny of chatbot advice, including allegations that ChatGPT encouraged a student who later died by suicide.
An OpenAI spokesperson told The New York Times that Wan’s death was an “incredibly heartbreaking situation” and said she had been using a version of ChatGPT that is no longer available. According to court documents, OpenAI retired the model on Feb. 13, one day before Wan’s family discovered her body.
Published: Oct 30, 2026 07:30 am