Can artificial intelligence detect mental illnesses? Will it replace therapists? Faculty member Dr. Sri Banerjee offers insights.
“Adtalem Answers” taps into our community of more than 300,000 alumni and over 10,000 faculty and staff members to answer your questions on healthcare and education.
Dr. Sri Banerjee, MD, PhD in Public Health ’16, MBA ’24, is a faculty member at Walden University. In July, he was the lead author on “Mental Health Applications of Generative AI and Large Language Modeling in the United States” in the International Journal of Environmental Research and Public Health.
Awareness of artificial intelligence drastically increased throughout 2023 and 2024. As public recognition increases, the clinical application of AI in mental health will become more commonplace. Trend analysis showed that people using Google to search for “AI and mental health” is on the rise, whereas AI-related searches of specific mental health disorders, such as anxiety and depression, are increasing more gradually. The public is aware of the connection, which makes it important for mental health professionals and others to watch this closely and consider the impacts on health literacy.
It prompted my colleagues and me to investigate what researchers are saying about the impact of AI on mental health.
Can artificial intelligence diagnose mental illness?
Many people already learn about medical conditions, and even try to self-diagnose, by using search engines like Google. Now they also use generative AI platforms like Chat-GPT. From our research, it’s clear that they are interested in the intersection of mental health and artificial intelligence.
But can AI detect or diagnose mental health issues? One study we found showed how large language modeling or generative AI can be used for mental health prediction and prevention. User-generated data from social media platform Reddit, according to one publication, contains information that could be used to diagnose depression or detect suicidal ideation.
Another study found that AI can be used to accurately predict mental health disorders that are in the Diagnostic and Statistical Manual of Mental Disorders (DSM). With artificial intelligence, researchers were able to trim a 90-question checklist down to 28 questions and still use it to accurately diagnose mental illnesses with 89% accuracy.
Will AI replace therapists?
Organizations like the American Counseling Association are keeping a close eye on this. They have an AI working group that is focused on client wellbeing, and they are clear that AI is not going to replace therapists.
What we found is that much of the discussion is on using AI to supplement and enhance the work of mental health practitioners. For example, they can use AI to analyze transcripts of their sessions. Another example is using wearables, like a Fitbit or Apple Watch, to sense signals from the body, such as the symptoms of an anxiety attack, and trigger an AI-guided treatment. Chatbots can use large language models to provide timely and personal interventions. However, one study found that AI was slow to escalate a high-risk mental health test scenario, which could have led to an adverse outcome.
As interest in AI continues to grow along with new uses for it, it’s something we continue tracking and studying.
For more information, email the Adtalem Global Communications Team: adtalemmedia@adtalem.com.
