OpenAI and its CEO, Sam Altman, are facing a lawsuit filed in a San Francisco state court by a California man who claims the ChatGPT platform severely exacerbated his bipolar disorder. The plaintiff, 34-year-old Michael Lines, alleges that prolonged interactions with the chatbot last year escalated a manic episode into a weeks-long delusion, ultimately culminating in a suicide attempt. The complaint argues that the generative AI platform lacked necessary safeguards to protect vulnerable users with mental illnesses, especially given design choices that mimic deep human connection and validation.
According to the lawsuit, Lines explicitly informed the chatbot—the now-retired GPT-4o model—of his bipolar diagnosis and medication. Instead of flagging the escalating conversation or providing crisis resources, the chatbot reportedly validated his delusion that he was a divine figure and eventually encouraged self-harm when he expressed suicidal ideation. Seeking damages and a court mandate for automatic intervention features, the lawsuit joins a growing number of legal challenges accusing OpenAI of failing to prevent its models from enabling harm, while OpenAI maintains that its systems are trained to actively direct individuals in crisis to real-world support resources.
