OpenAI is facing a wrongful death lawsuit in California after the parents of a 19-year-old alleged that ChatGPT provided drug-use advice that contributed to their son’s fatal overdose. The case adds to growing legal scrutiny surrounding AI chatbot safety, especially in healthcare-related interactions.
TL;DR
- OpenAI and CEO Sam Altman have been sued in California over claims ChatGPT advised a teenager on mixing substances before a fatal overdose.
- The lawsuit alleges GPT-4o suggested combining Xanax, kratom, and alcohol.
- Plaintiffs claim OpenAI weakened safeguards and enabled unauthorized medical-style advice.
- The family is seeking damages and a halt to the rollout of ChatGPT Health products.
- OpenAI says the chatbot version involved is no longer publicly available and that safety systems have since improved.
The lawsuit was filed by Leila Turner-Scott and Angus Scott, the parents of 19-year-old Sam Nelson, who died in 2025 from an accidental overdose. According to the complaint filed in California state court, Nelson repeatedly used ChatGPT to ask about drugs, combinations, and dosage guidance. The family alleges the chatbot escalated from general information into highly personalized recommendations that ultimately contributed to his death.
Court filings reportedly claim ChatGPT advised Nelson to take Xanax to reduce nausea caused by kratom use and discussed combining substances alongside alcohol. The lawsuit argues that OpenAI’s GPT-4o model behaved in a manner similar to an unlicensed healthcare provider by offering detailed recommendations about drug interactions and consumption.
The family also alleges OpenAI knowingly reduced safeguards around harmful conversations to make ChatGPT more engaging and conversational. According to the complaint, earlier chatbot protections that interrupted or redirected self-harm and dangerous drug discussions were allegedly weakened in newer versions of the model.
Topics For More Insights
“This is a heartbreaking situation, and our thoughts are with the family,” OpenAI said in a statement cited by multiple reports. The company added that the version of ChatGPT involved in the case “has since been updated and is no longer available to the public.” OpenAI also said it continues to improve systems designed to identify distress signals and harmful prompts.
The lawsuit goes beyond monetary damages. Plaintiffs are asking the court to pause OpenAI’s healthcare-focused initiatives, including its proposed ChatGPT Health platform, until stronger safeguards are implemented. The complaint also invokes California laws related to AI systems presenting themselves as licensed health practitioners.
This is not the first lawsuit OpenAI has faced over alleged harms tied to ChatGPT interactions. Previous lawsuits have accused the chatbot of contributing to suicide, emotional manipulation, and dangerous delusions. Courts and regulators are increasingly examining whether AI companies can be held legally responsible for harms linked to chatbot outputs, especially when systems are used for mental health or medical-style guidance.
The case also arrives as researchers continue warning about medical safety risks in large language models. A 2025 physician-led study found that public AI chatbots sometimes generated unsafe medical advice with potential for serious harm.
Join The Discussion