The article examines the application of artificial intelligence (AI) in mental health care, focusing on a survey conducted among community members (CMs) and mental health professionals (MHPs). Mental illnesses are leading causes of disability worldwide, yet less than half of affected individuals receive evidence-based treatment. AI is increasingly recognized as a potential solution to these challenges, as it can automate tasks and provide new forms of support. The study aimed to assess current AI usage rates and understand the perceived benefits, harms, and risks among CMs and MHPs. Two online surveys were conducted in Australia, collecting data on demographics, comfort with technology, attitudes toward AI, specific AI use cases, and reported benefits and harms. The final sample included 107 CMs and 86 MHPs.
General attitudes toward AI varied: CMs were neutral, while MHPs exhibited a more positive stance. Among CMs, 28% had used AI, primarily for quick support and as a personal therapy tool. In contrast, 43% of MHPs used AI, mainly for research and report writing. Most users found AI generally helpful, but 47% of CMs and 51% of MHPs reported specific harms or concerns. Future AI applications in mental health evoked mixed feelings: some respondents were excited about AI’s potential, while others worried about reduced human connection, ethical concerns, and data security. Respondents believed AI could improve mental health care by enhancing accessibility, reducing costs, offering personalization, and increasing work efficiency. However, concerns included the loss of human interaction, ethical dilemmas, privacy and regulatory issues, medical errors, potential misuse, and data security risks.
Integrating AI into mental health systems requires careful consideration of legal and ethical aspects, along with safeguards to mitigate potential harms. Further surveys are planned to monitor the adoption and acceptance of AI over time. Overall, the study highlights that AI tools are increasingly being used in mental health care, but their integration should be approached cautiously, balancing potential benefits with associated risks.
