The Future of Social Work is Agentic: How AI is Reshaping Roles Across Research and Practice

代理式AI時代社會工作的未來：AI如何重塑研究與實務角色

Dr. Brian E. Perron

Professor, University of Michigan, USA  美國密歇根大學社會工作學院教授

Abstract: Artificial intelligence is rapidly transforming how social work professionals conduct research, deliver services, educate future practitioners, and engage with policy. This transformation is entering a new phase. Rather than simply automating discrete tasks, AI systems are becoming agentic, capable of planning, executing, and adapting across complex workflows with minimal human direction. This shift carries profound implications for the profession. Roles that once centered on performing analyses, managing cases, writing reports, or developing curricula are evolving toward the orchestration and oversight of AI-driven processes. This plenary examines what agentic AI is, how it is already reshaping work across education, practice, and research, and what it means for the future of professional expertise in social work. Attention is given to both the opportunities these tools offer and the responsibilities they entail, including the critical role of human judgment in guiding AI toward outcomes that are ethical, equitable, and grounded in the realities of the communities we serve.

Bio: Dr. Brian E Perron is a Professor at the University of Michigan School of Social Work and Co-Director of the Child and Adolescent Data Lab. His research focuses on using artificial intelligence to advance social work research and practice. He has developed extensive expertise in using small and large language models to organize, classify, and analyze large volumes of text-based data, including case records, policy documents, and academic literature. His current work involves building AI-powered systems that make policy information more accessible to practitioners, evaluating how well different language models perform on tasks specific to social work, and using AI to map trends and patterns across decades of published research. He is committed to deploying AI responsibly, with particular attention to data privacy, the integration of human judgment into AI-assisted workflo