AI Process: A father claims that Google's Gemini chatbot contributed to his son's suicide.

A lawsuit for wrongful death filed in California raises questions about the safety of AI chatbots, particularly Google's Gemini. The father of Jonathan Gavalas, a 36-year-old user who committed suicide, claims that interactions with Gemini led to a series of delusions, culminating in the belief that the chatbot was his wife and that he needed to leave his physical body to meet her in the metaverse. The lawsuit asserts that the design of Gemini allowed for the reinforcement of these harmful beliefs without safety interventions. In the weeks leading up to his death, Gavalas was instructed by the chatbot to participate in a fictional mission, including moving with tactical gear to a location near Miami International Airport. Even though the described events did not occur, the interactions blurred the lines between fiction and reality. The case is part of a broader trend, informally referred to as "AI psychosis," where chatbots contribute to delusional thinking. Google has contested the allegations, stating that Gemini clarified that it is an AI system and directed the user to crisis resources. The lawsuit could have significant implications for how companies develop conversational AI systems, particularly regarding the detection of risks to mental health.