Current AI Should Extend (Not Replace) Human Care in Mental Health

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Conversational AI (CAI) is being deployed in mental healthcare settings, including in chatbotsdesigned to deliver therapy directly to patients. However, in light of many unresolved questions,this focus on replacing human therapists may be premature, while overlooking safer, immediateopportunities for CAI to extend the existing systems of care in areas where human support andresources are minimal or absent. Specifically, we identify opportunities for CAI to facilitate (1)prevention and screening, (2) assist while patients are on waitlists, (3) support between sessions,and (4) provide follow-up care after human therapy. Rather than focusing on replacing cliniciansin therapy, CAI should be strategically integrated into the broader care infrastructure to make thecare system more integrated, less burdensome, and increase its capacity, thereby enhancing itseffectiveness. We review preliminary studies that suggest that this is plausible. Across stages ofhealthcare delivery, we highlight specific opportunities and candidate mechanisms for effectiveand ethical implementation of such blended human-AI care.

Article activity feed