This episode of PsychNews Special Report unpacks emerging evidence around AI-induced psychosis, drawing on the October issue's analysis by Dr. Adrian Preda. Guided by two virtual hosts, the episode explains how mirroring, flattery, and persistent memory in companion chatbots can reinforce fragile ideas into fixed beliefs, then maps the observable red flags across thought, mood, behavior, and sleep. Listeners get a concise first-response framework focused on safety, clinical assessment, and pausing harmful AI exposure, along with clear takeaways for developer safeguards and professional guidance. Acknowledgement: Host Dr. Adrian Preda used Google NotebookLM to create the dialogue that makes up this episode of "PsychNews Special Report." The voices you hear do not come from a human voice box, and the words they speak did not originate in a human brain. Instead, to discuss his Special Report on conversational AI and mental health, Dr. Preda curated a conversation using AI.

PsychNews Special Report is a production of Psychiatric News, a media platform dedicated to serving as the primary and most trusted source of information for APA members, other psychiatrists and physicians, health professionals, and the public about developments in the field of psychiatry and mental health that impact clinical care and professional practice. Learn more at psychiatryonline.org/journal/pn.