We introduce a novel application of large language models (LLMs) in developing a virtual counselor capable of conducting motivational interviewing (MI) for alcohol use counseling. Access to effective counseling remains limited, particularly for substance abuse, and virtual agents offer a promising solution by leveraging LLM capabilities to simulate nuanced communication techniques inherent in MI. Our approach combines prompt engineering and integration into a user-friendly virtual platform to facilitate realistic, empathetic interactions. We evaluate the effectiveness of our virtual agent through a series of studies focusing on replicating MI techniques and human counselor dialog. Initial findings suggest that our LLM-powered virtual agent matches human counselors' empathetic and adaptive conversational skills, presenting a significant step forward in virtual health counseling and providing insights into the design and implementation of LLM-based therapeutic interactions.

When we talk to people, we choose how much personal information to disclose about ourselves. How much we reveal, both in terms of breadth and depth, depends on many factors. In a therapeutic context, it is considered desirable to self-disclose, yet people sometimes hold back negative personal experiences, due to fear of being judged or revealing their weaknesses. Researchers Takahisa Uchida and colleagues from Osaka University hypothesized that people may find it easier to disclose their negative feelings to a robot due to its lower social position. The findings of their preliminary research were presented at the 2017 Robot and Human Interactive Communication symposium in Lisbon in August.