Generative artificial intelligence (AI) technologies are proliferating and have the potential to offer engaging and scalable ways to support treatment offer between-session support for patients, and even fully-automated interventions, as well as therapist training. We will provide an overview of these technologies, focusing on large language model based tools (e.g., those based on models like GPT-5, Claude, or other language models). We will also discuss considerations for their development and evaluation, including user experience research, rigorous safety testing, and collaborating with patients and clinicians during development. We will share a framework for evaluation and key directions for establishing safety, efficacy, and equitable access to effective treatment. Finally, we will describe some use cases for LLM-based mental health applications.
Speakers
Elizabeth (Betsy) Stade is a computational clinical psychologist and research scientist at Stanford’s Institute for Human-Centered AI working in the Fidelity, Adaptation, Sustainability, and Training lab (PI: Shannon Wiltsey Stirman) and the Computational Psychology and Well-Being lab (PI: Johannes Eichstaedt) . Betsy studies how AI and large language models can be used for psychological assessment and treatment. She also uses natural language processing to measure and understand psychopathology, like depression and anxiety. Betsy did her graduate work at the University of Pennsylvania and her clinical residency at the VA Palo Alto Health Care System, and is a licensed clinical psychologist in California. Her research has been supported by the National Science Foundation.
Shannon Wiltsey Stirman is a Professor in the Department of Psychiatry and Behavioral Sciences at Stanford and a Psychologist at the National Center for PTSD’s Dissemination and Training Division. She is the co-director of the NIMH-funded CREATE Center, which focuses on using AI to support high quality evidence-based treatment for PTSD. Areas of research emphasis include implementation science (particularly training, fidelity, adaptation and sustainment), evidence-based treatment for PTSD, depression, and suicide prevention, and use of technology to support access to evidence-based mental health interventions. As a co-lead of the Department of Psychiatry and Behavioral Science’s Mental Health Innovation and Technology Hub, she worked with a team at Stanford to develop Pause a Moment, a digital wellbeing program for healthcare workers who experience COVID-19 stressors (pam.stanford.edu). She is the co-author of Getting Unstuck from PTSD: Using Cognitive Processing Therapy to Guide Your Recovery. Her research has been funded by the National Institutes of Health, VA QUERI, private foundations, and the Canadian Institute for Health Research.