Artificial Intelligence in Mental Health Practice: A Systematic Review of Applications and Relational Implications

This study presents a systematic literature review (SLR) following PRISMA guidelines to examine the integration of artificial intelligence (AI) in mental health practice, with a specific focus on its impact on therapeutic relationships, professional roles, and clients’ perceptions of empathy. A comprehensive search in Scopus using predefined keywords yielded 229 records, of which nine empirical studies met the eligibility criteria for synthesis. Findings reveal that AI applications, including empathic chatbots, therapists, client matching algorithms, machine learning–based telepsychotherapy, and digital monitoring tools, offer promising opportunities to expand access and provide emotional support. Evidence indicates that AI can convey signals of empathy and support therapeutic processes, yet its relational quality remains limited compared to human-mediated interactions. Professionals expressed ambivalence, viewing AI as both a facilitator of efficiency and a potential threat to autonomy and clinical integrity. From the client's perspective, AI systems were generally perceived as helpful and non-judgmental, though insufficient in managing complex cases such as suicide risk. Interpreted through the Technology Acceptance Model (TAM), adoption of AI is shaped by perceived usefulness and ease of use, underscoring the need for ethical safeguards, professional training, and responsible integration. Overall, AI should be positioned as a complementary tool that enhances rather than replaces human therapeutic practice.