The paper "Considering AI-driven therapy: When does human empathy matter?" by Matan Rubin, Hadar Arnon, Jonathan Huppert, and Anat Perry explores the role of empathy in the context of AI-driven therapy. The authors argue that while AI language models have shown promise in various therapeutic applications, such as psychoeducation and treatment adherence, human therapists remain indispensable in certain aspects of therapy. They define empathy as encompassing cognitive, emotional, and motivational dimensions and highlight that AI currently lacks the latter two. The paper discusses how these limitations affect the therapeutic process, particularly in building a strong working alliance and addressing patients' needs for human connection and empathy. The authors suggest that AI can be most effective in providing practical tools and support, while human therapists are crucial for fostering a genuine emotional bond and addressing more complex patient needs. They emphasize the importance of further research and dialogue to understand how AI and human therapists can best collaborate in mental health settings.The paper "Considering AI-driven therapy: When does human empathy matter?" by Matan Rubin, Hadar Arnon, Jonathan Huppert, and Anat Perry explores the role of empathy in the context of AI-driven therapy. The authors argue that while AI language models have shown promise in various therapeutic applications, such as psychoeducation and treatment adherence, human therapists remain indispensable in certain aspects of therapy. They define empathy as encompassing cognitive, emotional, and motivational dimensions and highlight that AI currently lacks the latter two. The paper discusses how these limitations affect the therapeutic process, particularly in building a strong working alliance and addressing patients' needs for human connection and empathy. The authors suggest that AI can be most effective in providing practical tools and support, while human therapists are crucial for fostering a genuine emotional bond and addressing more complex patient needs. They emphasize the importance of further research and dialogue to understand how AI and human therapists can best collaborate in mental health settings.