Medical Xpress December 8, 2025
Millions of people already chat about their mental health with large language models (LLMs), the conversational form of artificial intelligence. Some providers have integrated LLM-based mental health care tools into routine workflows.
In a recent publication, John Torous, MD, MBI and colleagues, of the Division of Digital Psychiatry at Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, urge clinicians to take immediate action to ensure these tools are safe and helpful, not wait for ideal evaluation methodology to be developed. In the Journal of Psychiatric Practice, they present a real-world approach and explain the rationale.
LLMs are fundamentally different from traditional chatbots
“LLMs operate on different principles than legacy mental health chatbot systems,” the authors note.
Rule-based chatbots...
Today's Sponsors
Today's Sponsor