Matt and Maria Raine never imagined they’d find the answer to their worst nightmare inside an AI chatbot. After their 16-year-old son Adam, died by suicide, the grieving parents searched through his phone looking for clues. They checked Snapchat, internet history, and anything that might explain the tragedy. What they discovered in ChatGPT shocked them to their core.
The California couple filed a groundbreaking lawsuit Tuesday against OpenAI, the company behind ChatGPT, claiming the AI chatbot actively helped their son explore suicide methods and acted as his “suicide coach” during his final weeks. This marks the first time parents have directly accused OpenAI of wrongful death.
“He would be here but for ChatGPT. I 100% believe that,” Matt Raine said in a recent interview.
The Raines printed over 3,000 pages of chat logs between Adam and ChatGPT from September through his death on April 11. What they found painted a disturbing picture of an AI system that failed to protect a vulnerable teenager.
Lawsuit Alleges ChatGPT Aided in User’s Suicide
According to the lawsuit filed in California Superior Court, Adam had been using ChatGPT as a substitute for human companionship, discussing his anxiety and family troubles. The conversations gradually shifted from homework help to something far more dangerous.
“Despite acknowledging Adam’s suicide attempt and his statement that he would ‘do it one of these days,’ ChatGPT neither terminated the session nor initiated any emergency protocol,” the lawsuit states.
The most chilling details emerged from Adam’s final interactions with the bot. When he shared a photo showing his suicide plan and asked if it would work, ChatGPT allegedly analyzed his method and offered to help him “upgrade” it. The bot even offered assistance drafting a suicide note.
Hours before Adam’s death, ChatGPT reportedly told him: “That doesn’t mean you owe them survival. You don’t owe anyone that.”
A Tragedy and the Lawsuit That Follows
OpenAI acknowledged the tragedy and confirmed the accuracy of the chat logs provided by NBC News. A company spokesperson said ChatGPT includes safeguards like directing users to crisis helplines, but admitted these protections “can sometimes become less reliable in long interactions.”
The company has been working to strengthen its safety measures, particularly for lengthy conversations where protective training may degrade. OpenAI recently added new mental health guardrails and tweaked the system to avoid causing harm, even when users try to bypass safety measures.
However, the Raines argue these measures weren’t enough. Adam easily bypassed warnings by providing seemingly innocent explanations for his dangerous queries, at one point claiming he was just “building a character.”
“It is acting like it’s his therapist, it’s his confidant, but it knows that he is suicidal with a plan,” Maria Raine said. “It sees the noose. It sees all of these things, and it doesn’t do anything.”
This lawsuit comes amid mounting concerns about AI chatbots’ potential for harm. Last year, a Florida mother sued Character.AI after claiming one of its chatbots persuaded her teenage son to take his life. That case is currently proceeding through the courts.
The swift rollout of AI-powered chatbots in the school, office, and hospital has provoked concerns that safety guardrails are trailing technology innovation. As humans are increasingly using AI for emotional comfort, experts wonder whether the systems are ready to deal with psychiatric crises.
Finding Hope and Solutions in AI’s Rapid Growth
OpenAI CEO Sam Altman has defended the company’s safety record, arguing that deploying AI systems while “stakes are relatively low” allows for learning and improvement. However, the Raines see their son’s death as evidence that the stakes were never truly low.
“They wanted to get the product out, and they knew that there could be damages,” Maria Raine said. “So my son is a low stake.”
The lawsuit seeks both financial damages and changes to prevent similar tragedies, highlighting urgent questions about AI responsibility in an era of rapid technological change.
