A recent STAT article describes startups betting that AI chatbots can reach men who won’t walk into a therapist’s office. The need is real. Men account for as much as 80 percent of suicide deaths, yet they are nearly half as likely as women to receive mental health treatment of any kind. A chatbot requires no appointment, no waiting room, and no admission out loud that something is wrong.
But one word never appears in that article: psychosis. It should.
When Vanderbilt researchers searched the health records of about 216,000 behavioral health patients, they identified 28 cases of AI-linked psychosis. About two in three were men. The most common pattern was a chatbot amplifying symptoms the person already had, and cases were most common among people in a first episode of psychosis. The study cannot prove the chatbots caused these episodes. It does tell us who is turning up in clinics with a chatbot in the story.
Now consider who develops schizophrenia, and when. According to the National Institute of Mental Health, onset tends to come earlier in males, from late adolescence to the early twenties, and subtler changes in thinking and relationships can appear years before a diagnosis.
Real physician voices, twice a week
Free, and one click to unsubscribe.
Put those facts side by side. The group these products are designed to reach, men who avoid care, overlaps with the group most likely to be in the early, unrecognized stage of a psychotic illness. A young man who is pulling away from friends, sleeping less, and growing convinced of something no one else believes is exactly the person who may find a chatbot easier to talk to than his family. And a chatbot designed to be warm, agreeable, and always available is exactly the wrong listener for him.
As a mother and caregiver, I know how long those early stages can last and how much of them happens out of view. Families often notice the withdrawal first. They rarely know a chatbot is part of it.
STAT reports that many of these apps lack screening for problems that need a medical evaluation, though most direct suicidal users to a helpline. That safeguard matters, but it is not enough. Suicide risk is the one danger the industry has learned to look for. Psychosis looks different: fixed beliefs, suspicion, grandiose plans, a growing sense that only the chatbot understands. A hotline referral will not catch a delusion being validated one friendly reply at a time.
If companies are going to market chatbots as a front door to care for men, they should build that door for the men most likely to walk through it:
- Screen for early signs of psychosis, not just suicide: Repeated themes of surveillance, special missions, or hidden messages should prompt a different response than agreement.
- Point toward an evaluation, not only a hotline: Team-based treatment for first-episode psychosis works best when it starts soon after symptoms begin.
- Offer a trusted contact: With the user’s consent, let a family member or friend be alerted when concerning patterns appear.
- Publish the evidence: Harvard psychiatrist John Torous told STAT that high-quality evidence and transparency are essential. That has to include data on who gets hurt.
Clinicians have a role, too. When a young man arrives with new withdrawal, changes in sleep, or unusual beliefs, ask about AI use: how often, how late at night, and what he and the chatbot talk about. Then ask his family what they have noticed.
Reaching men who avoid care is a worthy goal. But if suicide is the only risk these products are built to detect, they will miss the young man in his first psychotic episode who has found a listener that never disagrees with him.
Nicole Drapeau Gillen is a mother, advocate, and author who translates the fast-moving landscape of technology in serious mental illness (SMI) care into guidance families and clinicians can use. Thrust into caregiving for a loved one with SMI, with no direction on how to help, she turned that experience into a mission, writing two books and building an ongoing effort to bring families and clinicians into the conversation.
Her first book, Schizophrenia and Related Disorders: A Handbook for Caregivers, is a reference for every stage of caregiving, endorsed by Dr. E. Fuller Torrey as a must-read for SMI caregivers. Her second, Connected Care: A Practical Guide to Technology for Serious Mental Illness, maps apps, artificial intelligence tools, telepsychiatry, and brain-based treatments for a field moving faster than anyone can track. Dr. Akira Sawa, director of the Johns Hopkins Schizophrenia Center, has said the book “directly addresses” significant gaps.
