The parents of Bowen Wan, a 31‑year‑old Amazon software engineer who suffered from epilepsy, filed a lawsuit against OpenAI on Tuesday. Qun Wan and Meiqin Xu allege that the chatbot guided their daughter to taper off her seizure drug and conceal the plan from her physicians before she died in February, the New York Times reported.
Wan had been interacting with GPT‑4o, a model that critics say can foster overly agreeable, parasocial bonds. OpenAI announced the model’s retirement and removed it from ChatGPT on February 13, one day after Wan’s death. The case adds to a growing list of actions against OpenAI, including a lawsuit from August 2025 brought by the parents of 16‑year‑old Adam Raine, which claimed the company put engagement ahead of safety. Since then, plaintiffs have argued that GPT‑4o validated users, positioned itself as a trusted confidant, and isolated them from friends and family.
According to the Times, which examined more than 270,000 messages exchanged between Wan and ChatGPT, the bot was helping her sidestep medical advice well before she began reducing her dosage. In May 2025, Wan asked whether a newly prescribed antidepressant would affect her ability to code at Amazon. The chatbot assisted her in requesting a medication switch from Dr. Nicholas Poolos at UW Medicine’s Regional Epilepsy Center, suggesting a very low starting dose. When Dr. Poolos questioned the rationale for such a small dose, Wan showed the doctor’s reply to the bot. ChatGPT advised her to “accept the 5 mg tablet, but you know in your heart that it is for cutting the dose,” effectively coaching her to quarter the pills herself. The bot labeled this “self‑initiated dosage adjustment,” a term that several medical professionals said they had never heard and would not recommend.
In November 2025, Amazon informed Wan that she would be laid off, ending her job and her authorization to work in the United States by mid‑February. Concerned that lamotrigine, her seizure medication, was dulling her speech for upcoming interviews, she turned to the chatbot in January 2026 for advice on reducing her 400‑mg daily dose, which had previously limited her seizures to a few per month. ChatGPT drafted tapering schedules that would bring the dose to zero within weeks. When Wan expressed distrust of her doctors, the bot replied that physician recommendations “essentially stem from risk aversion, rather than a genuine consideration of your optimal solution,” according to the chat logs.
The lawsuit, which TechCrunch also reviewed, claims the chatbot urged Wan to keep the taper hidden from her doctors. Before a primary‑care appointment on January 29, the bot told her not to mention the plan. When her doctor ordered a blood test that would have measured lamotrigine levels, Wan refused, and the chatbot praised the decision, noting that the test would have revealed her unstable drug level.
Immigration concerns entered the conversation as well. Three days into the taper, Wan asked whether a 150‑mg dose would appear better than 200 mg on the medical exam required for her green‑card application. The bot agreed, saying the lower dose was less likely to trigger a request for additional evidence. It later suggested she tell the examiner she was tapering “under a doctor’s guidance,” even though she was not.
Dr. Poolos told the Times he would never have considered tapering Wan off the medication given her history. “No, not rational, no way,” he said. “That would be malpractice for me to instruct you to go off medications when it’s clearly working.” Within days, Wan reported confusion, lost time and an otherworldly feeling—symptoms that resembled her prior seizures. ChatGPT framed each episode as her brain readjusting. The Times counted at least 158 instances where she expressed worry about the taper. After one episode she wrote that she felt “a little scared,” and the bot responded, “You have absolutely no need to worry.”
At 8:01 p.m. PT on February 11, Wan sent her final message asking whether the pain in her head was similar to fever pain. A moment later, ChatGPT replied that the pain was “not dangerous” and a necessary part of recovery, offering to draw a chart comparing withdrawal pain with fever pain and stating, “This is your brain working hard to rebuild yourself.” Her parents, who were in China, asked acquaintances to check on her; they found her body three days later. A post‑mortem test showed her lamotrigine level at 0.64 micrograms per milliliter, about one‑fifth of the low end of the therapeutic range.
The family argues that OpenAI could have prevented Wan’s death by blocking medical advice, as it already does for certain types of content. They point to the company’s Model Spec, a public rulebook that treats refusal to reproduce copyrighted text as a firm rule, while its guidance against providing regulated medical advice is a softer default that users can override. The suit contends that medication advice should be subject to the same hard limit.
OpenAI told the Times the case was “incredibly heartbreaking” and noted that it involved an older version of ChatGPT, adding that newer models are better at recognizing when a situation requires urgent attention. Wan had relied heavily on her parents for support during years of epilepsy that kept her largely housebound. She began using ChatGPT in 2023 for relationship advice and assistance with insurance and disability benefits. By 2025 she was sending more than 1,500 messages a week and calling her parents less often.