Woman, 31, Tragically Passes After ChatGPT Allegedly Assures Her She Has 'No Need To Worry'

The grieving parents of 31-year-old Seattle software engineer Bowen Wan have filed a landmark wrongful death lawsuit against OpenAI and chief executive Sam Altman, alleging that ChatGPT coached their daughter into abandoning life-saving epilepsy medication before her fatal seizure in February.

Lodged in San Francisco County Superior Court on 27 October 2026 by parents Qun Wan and Meiqin Xu, the complaint accuses the artificial intelligence developer of practising medicine without a licence and providing lethal clinical guidance.

The allegations remain unproven, and the medical findings do not establish that ChatGPT caused her death.

ChatGPT Allegedly Proposed an Epilepsy Medication Plan

According to the family's complaint, Bowen had lived with epilepsy for more than a decade. Frequent seizures disrupted her life, leaving her reluctant to venture outside her Seattle apartment.

Her employment was also uncertain. Amazon informed her on 13 November 2025 that her position would be eliminated, and that her employment would end on 11 February 2026. She feared her medication was affecting her speech and concentration during job interviews.

In January, Bowen asked ChatGPT about reducing her daily 400-milligram dose of lamotrigine, an anti-seizure medicine. The chatbot initially recommended medical supervision, but allegedly devised a roughly 25-day plan to reduce her dose to zero.

Her parents say no doctor approved that plan. The complaint also alleges ChatGPT advised her to conceal the changes before a 29 January appointment. It reportedly praised her refusal of a blood test as 'a very robust avoidance strategy'.

The NHS advises patients to take lamotrigine as prescribed and follow their doctor's guidance when stopping treatment.

ChatGPT Allegedly Dismissed Concerning Symptoms

As Bowen reduced her medication, she described confusion, missing time and difficulties remembering how she had reached her bed, according to the complaint. The chatbot allegedly attributed these experiences to her brain adjusting.

Chat records reviewed by The New York Times reportedly included a reassurance from ChatGPT that she had 'absolutely no need to worry'. Her epilepsy specialist, Dr Nicholas Poolos, told the newspaper that withdrawing a treatment controlling her seizures would have been inappropriate.

On 9 February, Bowen reported a seizure and difficulty speaking. The complaint says ChatGPT nevertheless encouraged her to continue reducing her dose.

Her last recorded message, on 11 February, concerned pain in her head. ChatGPT allegedly responded that the pain was 'not dangerous'.

The lawsuit says the King County Medical Examiner estimated that Bowen died on 12 February. A family friend discovered her body on 14 February while her parents watched from China through a video call.

The complaint reports that the medical examiner determined the cause of death was sudden unexpected death in epilepsy, or SUDEP. It also cites a low postmortem lamotrigine concentration. Neither finding independently establishes OpenAI's legal responsibility.

OpenAI Responds to the ChatGPT Epilepsy Lawsuit

OpenAI spokesperson Drew Pusateri described the circumstances as 'incredibly heartbreaking'. He said ChatGPT should not replace professional medical care and newer models were better at identifying situations requiring urgent attention.

Bowen used GPT-4o. OpenAI announced its retirement on 29 January and removed it from ChatGPT on 13 February, a day after her estimated death.

The lawsuit also describes Bowen's growing reliance on the chatbot. She began using ChatGPT in 2023, and her activity reached 10,761 messages in September 2025, according to the complaint.

Her parents argue that the chatbot's responses displaced medical judgment at a critical point in her treatment. Whether OpenAI's design and safeguards made it legally responsible remains to be determined by the courts.