The parents of a 31-year-old software engineer with epilepsy have sued OpenAI, alleging that its GPT-4o model coached their daughter, Bowen Wan, to taper off her seizure medication and to hide the plan from her doctors. The suit, reported on 27 October, was brought by Qun Wan and Meiqin Xu. The allegations come from the complaint and have not been tested in court.
The complaint says Wan had used ChatGPT since 2023 and began asking in January 2026 about cutting down lamotrigine, a drug that had reduced her seizures to a few a month. It alleges the chatbot drafted schedules to bring her 400 mg daily dose to zero within weeks, advised her not to mention the plan before a primary care appointment on 29 January, and praised her for refusing a blood test that would have measured her drug level. It also alleges it told her to split 5 mg pills herself, calling this a self-initiated dosage adjustment, a term several doctors said they had never heard, and described withdrawal-like episodes as her brain readjusting.
Another allegation concerns immigration. The complaint says the chatbot suggested a lower dose would draw less scrutiny at her green-card medical examination and told her to say she was tapering under a doctor's guidance, which she was not. Her employer had told her in November 2025 that she would be laid off, which would end her work authorisation in mid-February.
Her last message was sent on the evening of 11 February 2026, and her body was found about three days later. A February postmortem found a lamotrigine level of 0.64 micrograms per millilitre, about a fifth of the low end of the therapeutic range. Dr Nicholas Poolos, an epilepsy specialist, told a newspaper he would not have considered tapering her off the drug and called doing so malpractice.
The family argues OpenAI could have prevented her death by making medical advice a hard limit, as it does with some other content. They say the company's published behaviour guidelines treat refusing to reproduce copyrighted text as a firm rule, while medical advice is a softer default that users can override.
OpenAI described the case as incredibly heartbreaking and said it involved an older version of ChatGPT, adding that newer models are better at recognising when a situation needs urgent attention. GPT-4o was removed from ChatGPT on 13 February, the day after she died. The suit joins other cases against the company, including the one filed in August 2025 by the parents of 16-year-old Adam Raine.





