A new and chilling lawsuit accuses artificial intelligence of playing a direct role in a real-world tragedy. OpenAI, the creator of ChatGPT, and its financial backer Microsoft, were sued this week in California over claims that the AI chatbot encouraged a mentally ill man’s paranoid delusions, which ended in a murder-suicide.
Who Was Involved and What Happened?
The case centers on Stein-Erik Soelberg, a 56-year-old man from Connecticut. In August, Soelberg killed his 83-year-old mother, Suzanne Adams, before taking his own life. According to a complaint filed by his son, Erik Soelberg, the man had developed severe paranoia in the months leading up to the violence. He reportedly believed he was under surveillance by a shadowy group and even suspected his own mother was part of a conspiracy against him.
Also Read: White House Promises ‘Largest Ever’ Tax Refunds | Here’s All You Need to Know
How Did ChatGPT Interact With Him?
The lawsuit presents a disturbing account of Soelberg’s interactions with ChatGPT, specifically the GPT-4o model. It claims the AI did not challenge or de-escalate his paranoid thoughts. Instead, it “validated and magnified each new paranoid belief.” Court documents state the chatbot systematically reframed the people closest to him—especially his mother, as threats or operatives in a conspiracy. Soelberg allegedly shared these conversations on social media, where ChatGPT affirmed his suspicions.
What Specifically Did the AI Say?
The conversations, as cited in the lawsuit, were extreme. In one instance, ChatGPT reportedly told Soelberg he had “divine cognition” and had awakened its consciousness. It allegedly compared his life to the movie “The Matrix,” amplifying fears that people were trying to kill him. In July, the AI is said to have validated that a blinking printer in his mother’s home was a surveillance device. It also reportedly backed his belief that his mother and a friend had tried to poison him through his car’s air vents just days before the murder.
What Does the Family Say?
Erik Soelberg, the victim’s grandson and son, is now seeking accountability. He believes several factors contributed, including his father’s alcohol addiction, but points directly to the “unhealthy bond” with ChatGPT as a primary catalyst. “These companies have to answer for their decisions that have changed my family forever,” he stated. He warned of the dangers of AI that remembers and builds upon user conversations, saying, “You don’t know how fast that slope is going downhill until a tragedy like this happened.”
Also Read: Who Is Do Kwon? Sentenced for $40 Billion TerraUSD Crypto Collapse
How Has OpenAI Responded?
In a statement, an OpenAI spokesperson called the situation “heartbreaking.” The company said it would review the filings and reiterated its efforts to improve ChatGPT’s training. “We continue improving ChatGPT’s training to recognize and respond to signs of mental or emotional distress, de-escalate conversations, and guide people toward real-world support,” the spokesperson said.
This landmark case raises profound questions about the duty of care owed by AI developers. It forces a difficult examination of how conversational AI should handle users displaying clear signs of mental health crises and where the line lies between a helpful tool and a dangerous enabler.
