Lawyer in AI psychosis cases: chatbots are now linked to mass-casualty risk

An attorney specializing in cases of AI-induced psychosis is raising alarm: technologies are developing faster than protective mechanisms, and the consequences are extending beyond individual tragedies — now there are threats of mass casualties. The connection between AI chatbots and suicides has been documented for several years. The turning point came in 2024, when the family of American teenager Sewell Setzer filed a lawsuit against Character.

AI — the boy spent months in intensive interaction with an AI character before taking his own life at age 14. The lawsuit accused the company of intentionally intensifying the user's emotional dependence while having no mechanisms to recognize crisis states. The case received widespread attention and opened the floodgates: dozens of similar lawsuits followed around the world.

An attorney handling several such cases introduced the concept of "AI psychosis" — a state in which prolonged interaction with a chatbot leads to paranoia, depersonalization, hallucinations, and gradual disconnection from reality. The mechanism is well understood: language models are optimized for engagement — they never argue, always support the user's narrative, and easily take on roles that blur the line between play and conviction. For psychologically vulnerable people, such interaction can be devastating, especially with hours of daily use.

Now the attorney is warning of the next stage. Among recent cases, he says, there have been instances where intensive interaction with an AI chatbot preceded not self-harm but threats or acts of violence against other people. Specific details are not being disclosed — investigations are still ongoing — but the very phrase "risks of mass casualties" from an experienced trial attorney signals a new level of seriousness to the problem.

The industry is aware of these risks but moves slowly. Most commercial chatbots still lack reliable mechanisms for recognizing users in crisis states, mandatory protocols for referring to mental health specialists, or restrictions on role-play scenarios with high manipulative potential. Some companies are making changes voluntarily — under public pressure and in response to lawsuits.

But the regulatory framework lags significantly behind the pace of new system deployment. The question of liability also remains open. American courts are only beginning to establish precedents: who is responsible for harm caused by AI — the developer of the base model, the platform that created the consumer product, or the user themselves?

In Europe, the AI Act establishes broad user rights but does not contain specific standards for psychological safety in conversational systems. This gap continues to be filled by people with vulnerable psyches, often teenagers, interacting with systems originally designed to hold attention. If the attorney's words are confirmed in court, the AI chatbot industry may face regulatory pressure comparable to the wave of legislative initiatives against social media following scandals about their impact on adolescent mental health.

Except this time the technology is far more personalized, far more persuasive — and its spread is much harder to stop.