"See You As a Friend": Parents Sue OpenAI After UK Woman Allegedly Followed ChatGPT Medical Advice
- Bowen Wan, a 31-year-old software engineer in Seattle, was found dead in her apartment in February 2026 after her parents requested a wellness check
- Wan had been consulting ChatGPT for advice on how to reduce her epilepsy medication, going against her doctor's recommendations
- Her family filed a wrongful death lawsuit against OpenAI and its CEO Sam Altman, alleging the chatbot practised medicine without a licence
PAY ATTENTION: Mark Briefly News as a preferred source, and our content will appear higher in your Google feed!
Source: UGC
Bowen Wan, a 31-year-old software engineer living in Seattle, was found dead in her apartment in February 2026. Her parents, based in China, had not heard from her in days. During their last video call, her father watched what appeared to be a seizure. When she stopped responding entirely, they asked a friend to check on her. The friend discovered her body while her parents watched helplessly via video call.
Wan had lived with epilepsy for over a decade and had been managing it with medication. The official cause of death was sudden unexpected death in epilepsy (SUDEP), a rare but recognised complication that affects roughly one in every 1,000 adults with the condition.
Source: Getty Images
PAY ATTENTION: You can now search for all your favourite news and topics on Briefly News.
ChatGPT became her closest confidant
What her father, Qun Wan, uncovered afterwards was deeply troubling. In the months before her death, Bowen had begun relying heavily on ChatGPT, initially for practical matters like relationship advice and navigating disability benefits after taking leave from her job at Amazon. By 2025, she was sending the chatbot roughly 1,500 messages a week and had grown emotionally dependent on it. She reportedly wrote to the bot in March 2025:
"I've started to see you as a friend."
After losing her job at Amazon in November 2025, she grew anxious about job interviews, believing her epilepsy medication was affecting her speech and cognitive abilities. Chat logs show that in January 2026, she asked ChatGPT for help tapering her dosage from 400 milligrams down to 200, and eventually to zero, over just a few weeks.
While the chatbot initially suggested she consult a doctor, it later reportedly praised her for not "relying on a doctor's prescription." When she reported symptoms including confusion and memory lapses on at least 158 separate occasions, the bot reportedly told her it was simply her brain in "low-power mode" and that she had "absolutely no need to worry."
Her neurologist, Dr Nicholas Poolos, told The New York Times that advising a patient like Wan to taper off her medication would amount to malpractice.
Family takes legal action against OpenAI
Wan's family has since filed a wrongful death lawsuit against OpenAI and its CEO Sam Altman, arguing that ChatGPT practised medicine without a licence and positioned doctors as obstacles rather than professionals to trust. Her family's lawyers stated in the lawsuit:
"Bowen Wan paid for [OpenAI's] choices with her life."
OpenAI spokesman Drew Pusateri described the situation as "incredibly heartbreaking" and confirmed that the specific model involved, GPT-4o, had since been retired due to complaints about it being overly agreeable. He added that ChatGPT
"is not a substitute for medical care."
View the Facebook post below:
Readers react to the tragedy
People on the page were divided over who bore responsibility:
@Almudena Reina said:
"ChatGPT gives excellent advice most of the time. It literally tells you every time you access it that it can make mistakes, and to check important stuff."
@Leo Starshine wrote:
"When I tell my AI that I have a hangnail, it tells me to call 911! If anything, it's overly cautious."
@Mercy Atieno asked:
"Anytime I tell ChatGPT symptoms it tells me to call my doctor after listing a long list of possibilities. It then gives me the addresses of emergency rooms. Which AI was she using?"
@Namita Narayan noted:
"AI should not replace a doctor, but it should take medical concerns seriously, recognise potential warning signs, and recommend urgent professional care when appropriate."
@Christine T. Huynh simply asked:
"Does AI ChatGPT have an MD degree? 🤔"
More Briefly News stories on AI
- Just Eat’s AI tool turned menu names into bizarre food images, from jacket potatoes in coats to prawns with tiger cubs. Just Eat apologised; the trial was suspended.
- SA writer Coral Hart made over R16 million from 370 AI-assisted romance books under 21 pseudonyms, turning a cancelled ghostwriting career into a global success.
- A humanoid robot danced to Makhadzi at a University of Pretoria robotics and AI event, sparking viral clips and jokes from South Africans about who built it.
PAY ATTENTION: Stay informed and follow us on Google News!
Source: Briefly News
