Artificial intelligence (AI) supported applications have become increasingly prevalent in health care practice, with mental health services being no exception. AI applications can be employed at various stages of mental health services and with different roles. This study aims to understand the potential advantages and disadvantages of using AI in mental health services, to explore its future roles, and outcomes through the opinions of mental health professionals engaged with AI. Thus, we conducted a qualitative study with semi-structured interviews with 13 mental health professionals who have expertise in AI, and a content analysis of the interview transcripts. We concluded that the use of AI in mental health services revealed advantages and disadvantages for clients, the profession itself, and experts. Our study emphasized four findings. Firstly, the participants were likely to have positive opinions about using AI in mental health services. Increased satisfaction, widespread availability of mental health services, reduced expert-driven problems, and workload were among the primary advantages. Secondly, the participants stated that AI could not replace a clinician but could serve a functional role as an assistant. However, thirdly, they were skeptical about the notion that AI would radically transform mental health services. Lastly, the participants expressed limited views on ethical and legal issues surrounding data ownership, the ‘black box’ problem, algorithmic bias, and discrimination. Although our research has limitations, we expect that AI will play an increasingly important role in mental health care services.