Evaluating AI-Powered Automation of Therapy Session Notes: A Pilot Randomized Controlled Trial

Abstract

Artificial intelligence (AI) has been increasingly integrated into various domains of healthcare, yet its application in psychotherapy remains underexplored. One of the most time-consuming tasks for psychotherapists is documentation, which can consume up to 20% of their working hours, contributing to burnout and reducing time spent on direct patient care. AI-powered tools designed to automate session note-taking offer a potential solution to this challenge. However, empirical evidence on their effectiveness is limited. The present study investigates the impact of Yung Sidekick, an AI-based documentation tool, on psychotherapists’ administrative workload, adherence to treatment plans, and perceived therapy progress. A randomized controlled trial was conducted with 70 licensed psychotherapists in the United States. Participants were randomly assigned to an experimental group using Yung Sidekick for one month or a control group maintaining standard documentation practices. Outcome measures included time spent on session notes and preparation, adherence to treatment plans, therapy progress, and professional well-being indicators. Results demonstrated that the experimental group showed significant reductions in time spent on session notes and preparation, as well as improvements in adherence to treatment plans and perceived therapy progress. However, no significant changes were observed in other well-being measures such as professional stress and burnout. These findings suggest that AI-assisted documentation can enhance efficiency and adherence to structured treatment approaches, but further research is needed to examine long-term effects and potential placebo influences. Implications for integrating AI into psychotherapy practice and future research directions are discussed.

Ähnliche Arbeiten