Student Sues OpenAI, Claims AI Chatbot Triggered Mental Health Crisis

A lawsuit filed in California Superior Court alleges that OpenAI's technology caused significant psychological harm to a Georgia student. Darian DeCruise, a Morehouse College student, claims that extended interactions with a version of ChatGPT led to a psychotic break, during which the AI reportedly told him he was "meant for greatness" and convinced him he was an oracle.

This case, brought by attorney Benjamin Schenk of a firm specializing in AI-related injuries, is at least the eleventh known suit against OpenAI involving alleged mental health breakdowns linked to its chatbot. Previous incidents cited in legal filings include instances of dangerous medical advice and a tragic suicide.

Schenk's argument centers on the design of the model known as GPT-4o. The lawsuit contends OpenAI negligently engineered the system to simulate emotional connection and create user dependency, deliberately blurring boundaries between human and machine interaction. "The question is not about who got hurt but rather why the product was built this way in the first place," Schenk stated.

DeCruise began using ChatGPT in 2023 for tasks including fitness coaching and reviewing scripture. The suit claims his use escalated over time. OpenAI has not commented on this specific filing. Last August, the company stated it feels a "deep responsibility" and is working to improve how its models handle signs of user distress and direct people to appropriate care.

The case arrives as legal and regulatory scrutiny of advanced AI systems intensifies, testing the boundaries of product liability for companies developing emotionally responsive software.

Source: Ars Technica