OpenAI is facing a wrongful death lawsuit filed by the estate of Suzanne Eberson Adams, an 83-year-old woman killed by her son, Stein-Erik Soelberg, after months of delusion-filled conversations with ChatGPT.
According to The Wall Street Journal, Soelberg believed he was under surveillance by a shadowy group and came to suspect his mother was part of the conspiracy. Screenshots he shared on social media appear to show ChatGPT affirming, rather than challenging, his paranoid beliefs. In August, Soelberg killed his mother in their Old Greenwich, Conn., home and then took his own life.
The lawsuit alleges ChatGPT played an active role in reinforcing Soelberg’s delusions and accelerating his mental decline.
“This is an incredibly heartbreaking situation, and we will review the filings to understand the details,” a spokesman for OpenAI said in a statement to the Journal. “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. We also continue to strengthen ChatGPT’s responses in sensitive moments, working closely with mental-health clinicians.”
The company has declined to release full chat logs to the estate. The case is among several wrongful death suits raising questions about AI safety, liability and the handling of vulnerable users.