Advances in Research on the Effectiveness of Artificial Intelligence-Based Cognitive Behavioral Therapy Chatbots for Alleviating Anxiety and Depression Symptoms: A Literature Review

Abstract

In recent years, advances in artificial intelligence (AI) have accelerated the development of digital mental health services, and cognitive behavioral therapy (CBT)-based chatbots have become an increasingly studied approach for anxiety and depression. These systems can deliver structured psychological support with relatively low cost, broad accessibility, continuous availability, and immediate interaction. This literature review synthesizes evidence from randomized controlled trials (RCTs), systematic reviews, and meta-analyses evaluating AI CBT chatbots for anxiety and depression. Current evidence suggests that these interventions can produce short-term reductions in mild-to-moderate depressive and anxiety symptoms, although effects are generally small to moderate and are more consistent for depression. However, confidence in these findings is limited by small or selective samples, short intervention and follow-up periods, substantial variation in chatbot design and control conditions, and limited evidence of long-term clinical benefit. Important concerns also remain regarding privacy, data governance, cultural and algorithmic bias, crisis recognition, and the reliability of AI-generated responses. Overall, AI CBT chatbots appear most defensible as accessible self-help or adjunctive tools rather than replacements for therapist-led CBT. Future research should prioritize larger and more diverse clinical samples, active comparison conditions, longer follow-up, transparent safety reporting, and carefully supervised human-AI models of care.