This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:

Conversational AI chatbots are a relatively new technology, but they are now ubiquitous, embedded in everything from customer service platforms to personal productivity tools. Such is their rapid adoption that potential negative consequences are yet to be fully understood.

Previous research has linked chatbot interactions to psychosis, a condition that can involve delusions, hallucinations or difficulty distinguishing reality from imagined experiences.

Much of this evidence was based on isolated case reports, simulated chatbot conversations and early observational research, and lacked large-scale clinical evidence.

But a new observational study published in JAMA Psychiatry reports that while cases of AI-related psychosis are rare, chatbot interactions may worsen symptoms for vulnerable individuals, particularly those experiencing a first episode of psychosis.

Researchers led by Zachary Bergson at Vanderbilt University Medical Center reviewed electronic health records covering approximately 216,000 behavioral health patients at the institution between December 2022 and April 2026. They used keyword searches and manual screening to look for patients with a documented psychotic disorder diagnosis alongside mentions of AI.

The search narrowed down the field to 73 individuals. Reviewers then read the medical notes to sort the cases into three categories. These were chatbot use that worsened symptoms, neutral chatbot use and delusional beliefs about AI without documented chatbot use. Finally, the team compared clinical background details among these patient groups.

From this group of 73, the study authors identified 28 patients whose records indicated that chatbot use worsened psychosis.

Source link