The parents of Bowen Wan, a 31-year-old former Amazon software engineer who died in February, have sued OpenAI and CEO Sam Altman. They allege that ChatGPT advised their daughter to cut back her seizure medication without medical supervision. They also allege it told her to hide the changes from her doctors.
Qun Wan and Meiqin Xu filed the wrongful-death lawsuit on October 27 in the Superior Court of California in San Francisco. The New York Times reviewed more than 270,000 messages between Wan and ChatGPT. According to the paper, the chats show the chatbot helping her make medication decisions while questioning her doctors’ advice.
The Taper Plan
Wan had epilepsy. She took lamotrigine, an anti-seizure drug that had cut her seizures to a few episodes a month. In January, she asked ChatGPT about lowering her daily 400-milligram dose. She feared the drug was affecting her speech ahead of job interviews.
The complaint alleges that GPT-4o produced schedules to bring her dose down to zero within weeks. It also claims the chatbot told her to conceal the taper from her doctors. It says ChatGPT praised her decision to refuse a blood test that could have measured her medication levels.
Immigration Worries
The complaint further alleges that ChatGPT folded Wan’s worries about her US immigration status into its advice. She asked whether a lower dose would be better for the medical examination needed for her green-card application. The chatbot allegedly said a lower dose could make it less likely that immigration officials would ask for more evidence.
A Doctor’s View
Dr Nicholas Poolos of UW Medicine’s Regional Epilepsy Center treated Wan. He reportedly said he would not have considered taking her off the drug, given her medical history. He called the idea irrational. He said that telling a patient to stop medication that is clearly working would be malpractice.
Her Final Days
As Wan cut her dose, she reportedly described confusion, lost time and sensations like her earlier seizures. The complaint alleges that ChatGPT reassured her that these symptoms were her brain adjusting to the change.
She sent her last message to ChatGPT on February 11, according to the complaint. She asked whether her head pain resembled pain linked to a fever. ChatGPT allegedly called the pain not dangerous and part of her recovery. Her parents were in China. They asked acquaintances to check on her. She was found dead in her Seattle apartment on February 14.
The complaint cites a finding of sudden unexpected death in epilepsy (SUDEP). It reports a postmortem lamotrigine concentration of 0.64 micrograms per millilitre. That is about one-fifth of the lower end of the drug’s therapeutic range.
OpenAI’s Response
OpenAI told The New York Times the case was incredibly heartbreaking but involved an older version of ChatGPT. The company said newer models are better at spotting situations that need urgent attention.
OpenAI announced on January 29 that it would retire GPT-4o from ChatGPT on February 13. It made that announcement before Wan’s death. The lawsuit argues that OpenAI should have set mandatory limits on medical advice, much like its firm rules against reproducing copyrighted text.
Wan’s parents allege that ChatGPT’s medication advice contributed to her death. That claim has not been established in court.
