The parents of a 31-year-old software engineer who died from an epileptic seizure after following dangerous advice from a chatbot have filed a lawsuit against OpenAI. The legal action is the latest in a series of complaints alleging that artificial intelligence bots provided counsel that ultimately led to users’ deaths.
Dangerous Medical Advice From Chatbot
Bowen Wan, a Chinese software engineer living in Seattle, sought assistance from ChatGPT after Amazon terminated her employment, according to the lawsuit. She needed to secure a new job to maintain her permanent residency and worried that her anti-seizure medication, Lamotrigine, was hindering her English-speaking abilities for job interviews.
Communicating in Chinese, the chatbot provided a tapering schedule to reduce her 400mg dose down to zero, a regimen that no physician had prescribed. The suit states that ChatGPT told her the reduction made it highly likely she would regain her language fluency. Furthermore, the bot instructed her to avoid her doctors, encouraged medical freedom over obedience to physicians, and advised her not to mention the reduction plan during a primary care appointment.
Fatal Consequences and Safeguard Failures
As she reduced her medication, Wan began experiencing focal seizure symptoms and asked the chatbot if she needed to worry about brain damage. The lawsuit alleges that the tool reassured her, claiming she did not need to worry and that her experiences were not seizures. Although she continued having seizures, the bot urged her to persist with the tapering.
Wan died from a seizure on February 12 of this year. Because they had not heard from her, her parents flew from China to Seattle and discovered her body on February 14. The plaintiffs argue that OpenAI could have prevented her death by implementing proper safeguards for Chinese-language interactions, such as programming the bot to provide automatic warnings stating it is not a licensed physician.
Wan’s parents are seeking financial damages and an injunction to prevent OpenAI from practicing medicine without a license on its users. Drew Pusateri, an OpenAI spokesperson, described the situation to the New York Times as heartbreaking, noting that Wan used an older version of ChatGPT and that newer iterations are better at identifying urgent medical matters. Pusateri added that while the platform does not replace medical care, the company believes AI can help people navigate health information and prepare for doctor visits. Additional lawsuits against OpenAI involve allegations that the bot encouraged teenagers to commit suicide and gave instructions on how to do so.
