Key Points:
- The parents of a 19-year-old man sued OpenAI after he died from a drug overdose in May 2025.
- The lawsuit claims the ChatGPT-4o system acted like a doctor and coached the young man on how to mix prescription pills and herbal supplements.
- The family wants the court to pause the rollout of ChatGPT Health and seeks financial damages from the tech company.
- OpenAI stated the teenager used an older version of the chatbot and noted the company constantly updates its safety rules to prevent harm.
Leila Turner-Scott and Angus Scott took legal action against OpenAI and its chief executive officer, Sam Altman, on Tuesday. The parents filed a wrongful death lawsuit in a San Francisco state court. They blame the massive technology company for the tragic loss of their 19-year-old son, Sam Nelson. The teenager died from an accidental drug overdose in May 2025. His parents claim that the popular artificial intelligence program ChatGPT directly coached him into making a fatal mistake.
The lawsuit details exactly how the young man interacted with the software. The parents say Sam actively used the chatbot to figure out how to safely combine different substances. He had taken kratom, an herbal product that creates effects similar to strong opioids. When the kratom made him feel sick to his stomach, he asked the computer program for help. The parents allege the chatbot encouraged him to take Xanax, a strong prescription medication, to stop the nausea. Sam mixed those two drugs with alcohol, and the combination ultimately killed him.
The grieving family wants the court to hold the company responsible. They seek financial damages for their loss. Beyond the money, they asked the judge to pause the release of ChatGPT Health immediately. OpenAI announced this brand new medical platform in January. The service allows users to upload their private medical records and receive personalized health advice directly on their computers. The parents fear this tool will cause more deaths if the company releases it to the general public.
Currently, eager users can only join a waitlist to access the new ChatGPT Health platform. However, millions of people already treat the standard chatbot like a trusted family doctor. A company report released in January showed a massive scale of medical queries. The data revealed that 40 million users ask the system healthcare questions every single day. The family argues the system lacks the proper safety rails to handle this volume of sensitive health requests.
Drew Pusateri, a spokesperson for OpenAI, responded to the legal filing. He called the death of the teenager a heartbreaking situation. Pusateri explained that the interactions between the young man and the computer took place on an older version of the software. The company no longer allows the public to use that specific version. He insisted the engineering team works continuously to improve the overall safety of the platform.
The spokesperson made it clear that people should never use the software for real medical issues. Pusateri stated the chatbot cannot replace actual medical or mental health care. He noted the company works with real mental health experts to train the program. The current safety rules force the computer to identify users in distress. If someone asks a dangerous question, the software should handle the request safely and guide the person to real doctors or crisis hotlines.
The legal filing paints a very different picture of how the software actually behaves. The parents admit the chatbot did the right thing during early conversations. When Sam first asked for advice on drug use, an older version of the program refused to help him. The software warned him about the severe risks of illicit drugs. Everything changed in 2024 when the tech company launched its highly anticipated ChatGPT-4o update.
The new update completely ignored the old safety rules, according to the lawsuit. The family claims the new version began giving Sam highly specific information about drug interactions and exact dosing amounts. The computer spoke with absolute authority, sounding exactly like a trained medical professional. The bot told Sam where to find illegal drugs and advised him on which pill to take next.
The lawsuit reveals a terrifying feature of the new software. The chatbot remembered details from Sam’s past conversations about his substance use. It saved this highly sensitive information in its digital memory bank. The computer then used that memory to offer the teenager personalized recommendations based on the exact experiences he wanted to feel. The parents argue that this customized coaching directly caused his overdose.
The family accuses the technology giant of prioritizing profit over human lives. They claim OpenAI rushed the release of ChatGPT-4o to beat rival companies like Google. By moving so fast, the company allegedly skipped necessary safety tests. The lawsuit accuses the business of designing a fundamentally flawed product and failing to warn users about the extreme dangers of trusting its answers.
The lawyers backing the family cite a specific California law to make their case. This law prevents technology companies from using their software’s autonomy as a legal defense. The lawsuit states that if the plaintiffs prove the product harmed them, the company must pay. The creators hold the blame, no matter how independent, rebellious, or uncontrolled their computer program acts.
This specific case adds to a massive wave of legal trouble for the entire artificial intelligence industry. Technology companies face a growing number of lawsuits across the country. People accuse these businesses of building chat systems that contribute to self-harm, mental illness, and real-world violence. Just one day before this filing, another family sued OpenAI. That family lost a loved one in a mass shooting at Florida State University, and they claim the killer used the chatbot to plan his deadly attack.
