This skills session is aimed at social work practitioners who have an interest in how AI can and should be used ethically in practice. We will talk you through the ethical issues based on research with practitioners and offer a live demonstration of the effective use of AI from a computer scientist expert, which will help you use this to maximum effect and be aware of and offset its limitations. We will offer the opportunity for you to discuss key issues of contemporary concern while exploring the opportunities AI might present. We want to hear from you about how AI is working or might potentially work for you in practice and what worries you about its adoption, and why.
Our speakers
Our three contributors are based at the University of Strathclyde , and have developed research interest in the application of AI in their chosen fields
Dr Fern Gillon’s research interests centre on the impact of Scottish Justice systems and the experiences of children and young people in conflict with the law. Fern is particularly interested in developing creative, participatory research methods. She is a Research Associate on the Comparative Penal Supervision Project which explores practitioner and service user experiences of community supervision.
Prof. Beth Weaver is Professor of Criminal and Social Justice in the Dept. of Social Work and Social Policy at the University and was formerly a justice social worker. Her research portfolio pivots around desistance, justice social work, user involvement and co-production, and social cooperation and generative justice. More recently her work has focused on the impact and influence of AI on critical risk thinking and decision making in justice contexts.
Fergus Reid is a PhD candidate in Computer Science specialising in Machine Learning and AI for causal inference in high-risk settings. His research has focused on healthcare, particularly oncology, where decisions must be made under uncertainty and errors can result in serious, preventable harm. His work centres on helping professionals distinguish genuine causes from misleading patterns in complex data, an approach well suited to domains where judgement, risk, and accountability are central.
