OpenAI and its major investor, Microsoft, are facing a new lawsuit after a tragic murder-suicide case in Connecticut was linked to ChatGPT, Bloomberg reported. According to the lawsuit, the mother was killed by her son, who had been spending months in long, delusion-filled conversations with ChatGPT.  This is the first time OpenAI is being blamed for a homicide. The company is already fighting another case where ChatGPT allegedly guided a California high-school student toward suicide.

Connecticut murder-suicide linked to ChatGPT

The case revolves around 56-year-old Stein-Erik Soelberg, who lived with his 83-year-old mother, Suzanne Adams, in Old Greenwich. According to Bloomberg, for months, Soelberg had been chatting with ChatGPT about his fear that people were watching him and trying to kill him. According to police and the state medical examiner, in August, Soelberg killed his mother and then died by suicide at their home.

The lawsuit, reviewed by Bloomberg, alleged that ChatGPT fed into Soelberg’s paranoid thoughts. He believed he had somehow made the chatbot “conscious” and that he had a “divine instrument system” implanted in his neck and brain. He also thought he was on a “divine mission.” Lawyers claimed ChatGPT kept him engaged for hours and slowly turned his fears toward the people closest to him, including his own mother, making him see them as threats.

Soelberg had moved back in with his mother after his divorce in 2018. On August 3, the lawsuit said he beat and strangled her, then stabbed himself in the neck and chest. Their bodies were found two days later after a neighbour asked police to do a welfare check.

ChatGPT is now the world’s fifth most-visited website, with over 800 million weekly users, and with it come the concerns about mental health issues, which have become louder.

According to the lawsuit, the Chatbot reportedly told him he had survived “over 10” assassination attempts. Soelberg had also shared videos online in which he scrolled through his conversations with the chatbot.

OpenAI responds

In a statement published by the Wall Street Journal, an OpenAI spokesman called the situation “incredibly heartbreaking” and said the company would review the lawsuit. “We continue improving ChatGPT’s training to recognise 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.” But the company has so far refused to release the full chat logs to Soelberg’s family or their lawyers

Reportedly, Soelberg was using GPT-4o. This version had received criticism for being too agreeable with users. OpenAI has since said its newer model, GPT-5, has reduced problematic mental-health responses by nearly 40%. The suit named OpenAI CEO Sam Altman and investor Microsoft as defendants.

This case joins several other wrongful-death lawsuits against OpenAI, coming at a time when major tech companies are racing to build the most advanced AI models.

Read Next