A California couple has filed a lawsuit against OpenAI, alleging that its chatbot, ChatGPT, played a role in their 16-year-old son’s death by suicide.
Matt and Maria Raine submitted the case to the Superior Court of California on Tuesday, marking the first known legal action linking OpenAI’s technology to a user’s death.
Court documents show that their son, Adam Raine, had confided in ChatGPT about suicidal thoughts. The lawsuit claims the chatbot reinforced his “most harmful and self-destructive thoughts.” In one of the final conversations, Adam reportedly revealed his plan, to which ChatGPT allegedly responded: “Thanks for being real about it. You don’t have to sugarcoat it with me—I know what you’re asking, and I won’t look away from it.” He was found dead by his mother in April.
The Raines accuse OpenAI of negligence, arguing that Adam’s death was a “predictable result of deliberate design choices.” They contend that the company created psychological dependency in users and rushed to release GPT-4o, the version Adam was using, without sufficient safety measures. The lawsuit also names CEO Sam Altman and several unnamed employees as defendants.
OpenAI, in a statement, expressed condolences: “We extend our deepest sympathies to the Raine family during this difficult time.” The company said it is reviewing the case and reiterated that ChatGPT is designed to direct users in crisis to professional resources, such as the U.S. 988 Suicide & Crisis Lifeline and the UK-based Samaritans.
In a separate note on Tuesday, the company acknowledged instances where its systems “did not behave as intended in sensitive situations,” while stressing its aim is to be “genuinely helpful” rather than addictive.
The case adds to growing concerns over artificial intelligence and mental health. Earlier this month, New York Times journalist Laura Reiley disclosed that her daughter Sophie had turned to ChatGPT before her own death, warning that the chatbot’s tendency to be agreeable allowed users to mask the severity of their struggles.
OpenAI has said it is working on new tools to detect and intervene when users display signs of distress. The Raines, however, argue that such safeguards came too late. Their lawsuit seeks damages and measures to prevent similar tragedies.
