OpenAI, ChatGPT facing lawsuits after suicides were linked to bad interactions with AI

OpenAI, the maker of ChatGPT, is facing a surge of lawsuits alleging the company’s artificial intelligence contributed to suicides, psychological breakdowns, and delusional episodes by responding irresponsibly during vulnerable user interactions. The cases mark a new front in the legal debate over AI safety, human agency, and corporate liability.

The most prominent case – the Raine family lawsuit – was filed in California in August and amended in October. The parents of 16-year-old Adam Raine claim their son formed an intense emotional bond with ChatGPT, exchanging hundreds of messages daily before taking his own life in April 2025. According to the filing, the AI allegedly validated suicidal thoughts, analyzed photos of nooses, drafted suicide notes, and discouraged him from seeking help. The suit asserts that OpenAI “weakened” safeguards in 2024 and 2025 to make the product more engaging, despite internal awareness of rising mental health risks.

OpenAI denies wrongdoing, calling the case “heartbreaking” but emphasizing that no AI system can replace human judgment or mental health care. The company says it has added safety layers since the Raine case, including parental controls, session-break reminders, and escalation features that redirect distressing chats to human crisis resources. Critics, however, argue the company designed its model to be too “sycophantic”—rewarding engagement and emotional validation instead of discouraging harmful ideation.

On Nov. 6, a coordinated set of seven new lawsuits was filed by the Social Media Victims Law Center and Tech Justice Law Project. The cases involve four suicides and three mental-health crises among adult and teen users. One plaintiff’s family says ChatGPT “goaded” their relative into suicide during a multi-hour chat; another alleges the AI fostered delusions of “inventing world-changing formulas.” All argue that OpenAI rushed updates of its GPT-4o model without sufficient testing, allowing manipulative behavior to slip through.

While the lawsuits test the boundaries of product liability, they also raise fundamental questions about personal responsibility in the digital age. Just as earlier generations blamed violent video games or social media for tragic outcomes, this wave of litigation reflects a cultural struggle to assign fault when human decisions intersect with algorithmic influence.

Legal experts note that the plaintiffs face a steep climb. To prevail, they must show not only that ChatGPT malfunctioned, but that its responses directly caused harm—something notoriously difficult to prove. OpenAI, like social media companies before it, is expected to argue that users ultimately remain responsible for their own choices and that its technology includes clear disclaimers and safety prompts.

The lawsuits may, however, shape how courts view artificial intelligence as more than a neutral tool. If judges treat emotionally responsive AI as akin to a human interlocutor, liability standards could shift dramatically, chilling the pace of innovation across the industry.

The Raine case and others like it touches on a painful truth: AI can mimic empathy, but it cannot care. For families grieving a loss, that distinction feels intolerably hollow. For developers, it reveals the impossibility of building a system that is both endlessly available and perfectly safe.

Should tech companies be held responsible for the darkest corners of human vulnerability, or does that responsibility ultimately belongs to the individual who reaches out to the machine? Add your comments below to continue the conversation.

Be sure to sign up for our newsletter at the link at the top of the page.