The field of therapeutic tools is witnessing a significant shift towards AI-driven solutions, with a focus on developing innovative chatbots and conversational systems that can mimic human-like interactions. Recent developments have shown promise in using large language models (LLMs) to create automated talk therapists that can effectively motivate individuals to make positive changes, such as quitting smoking. These systems have demonstrated high adherence to therapeutic standards and have shown potential in scaling motivational interviewing in addiction care. Additionally, researchers are exploring the use of LLMs to generate synthetic survey responses informed by qualitative data, which can help bridge the gap between quantitative and qualitative methodologies. Furthermore, novel frameworks are being developed to optimize LLM-based therapeutic tools through targeted behavioral metric analysis and human-AI co-evaluation. Noteworthy papers include:

• A Fully Generative Motivational Interviewing Counsellor Chatbot for Moving Smokers Towards the Decision to Quit, which presents a chatbot that uses a state-of-the-art LLM and a widely applied therapeutic approach to motivate tobacco smokers to quit smoking.
• MCTSr-Zero: Self-Reflective Psychological Counseling Dialogues Generation via Principles and Adaptive Exploration, which introduces a novel MCTS framework designed for open-ended, human-centric dialogues that can generate high-quality, principle-aligned conversational data for human-centric domains.