OpenAI is facing a lawsuit in California over allegations that its chatbot technology enabled the escalation of harassment by a user against his former partner.
The case, filed in a California Superior Court in San Francisco, involves a 53-year-old entrepreneur who allegedly developed delusional beliefs after prolonged interactions with ChatGPT and later used the tool to stalk and harass his ex-girlfriend.
The plaintiff, identified as Jane Doe, is seeking punitive damages and has requested a restraining order compelling OpenAI to block the user’s access, prevent new account creation, and preserve chat records for legal proceedings.
According to the lawsuit, OpenAI had received multiple warnings about the user’s behaviour, including an internal flag related to potential violent activity, but allegedly failed to take sufficient action.
The company has since suspended the user’s account but has not agreed to additional demands, including providing further information about the user’s activities.
The case adds to growing scrutiny over the real-world impact of generative AI systems, particularly concerns about their potential to reinforce harmful behaviour or delusional thinking.
The lawsuit was filed by Edelson PC, a law firm that has previously pursued cases involving alleged harm linked to AI systems, including incidents tied to chatbots developed by both OpenAI and Google.
OpenAI has not publicly commented on the case.
