Chatbots In Social Work Practice: A Scoping Review of Applications, Technologies, and Outcomes

Background and Purpose: Advancements in artificial intelligence are transforming behavioral health practice by enhancing accessibility and improving mental health outcomes. Chatbots, in particular, have shown effectiveness in general healthcare contexts. However, their application across diverse social work practice settings remains underexplored. Much of the existing literature discusses AI technologies in broad terms, highlighting a gap in understanding the design, implementation, and use of conversational interfaces tailored to social work populations. To address this gap, this scoping review examines the current state of research on chatbot use in social work, focusing on study characteristics, practice areas, target users, purposes, types of technologies, and reported outcomes.

Methods: The scoping review was conducted following the six-stages of Joanna Briggs Institute (JBI) guidelines: (1) identifying research questions, (2) identifying relevant studies, (3) study selection, (4) charting the data, (5) collating, summarizing, and reporting the results, and (6) consultation. A comprehensive search was performed in September 2025 across four major databases: Scopus, CINAHL, APA, PsycInfo, and Sociology Source Ultimate. Two reviewers screened the literature, applying the inclusion and exclusion criteria at both the title/abstract and full-text review stages, resolving any discrepancies through consensus. A total of 20 studies met the inclusion criteria. Data were extracted and analyzed using a narrative synthesis approach, and studies were categorized by research design: conceptual papers (n=5), intervention studies (n=5), and non-intervention data collection studies (n=10). The study selection and data charting were performed using Covidence.

Results: The findings indicate a rapid expansion of the field, with most studies published in 2025 (n=12) and led by social work scholars (n=11). Research is concentrated in the United States, with a subset of studies receiving federal funding (n=5). Mental and behavioral health emerged as the most consistent practice area across all research designs. The primary purpose of use is providing basic information and guidance, for which generative chatbots (e.g., ChatGPT, n=13) are most frequently utilized. In terms of users and purposes, intervention studies primarily focused on direct client support (e.g., intake and screening), whereas conceptual and non-intervention studies predominantly targeted social workers and educators for professional development and administrative automation. Empirical outcomes highlighted the potential of chatbots to improve clinical symptoms and optimize professional workflows.

Conclusion: This review highlights the rapid expansion of AI in social work, particularly within mental and behavioral health settings. Chatbots show promise in client screening and workflow efficiency; however, their integration into practice remains at an early stage. Evidence suggests these technologies function primarily as augmentative tools rather than substitutes for professional expertise. Future development should prioritize domain-specific chatbot systems grounded in social work theory and ethics. At the same time, attention to digital equity is critical to ensure that emerging technologies do not exacerbate existing disparities in access to care. Overall, a human-centered approach is essential to ensure that AI technologies strengthen, rather than compromise, the relational core of social work practice.

전반적으로 AI 기술이 사회 복지 실천의 관계형 핵심을 타협하는 것이 아니라 강화하도록 보장하려면 인간 중심 접근 방식이 필수적입니다.

Back to: Poster Presentations 8

<< Previous Abstract | Next Abstract >>