Enhancing Child Welfare Research and Translation through AI

by Dr. Morgan Cooley, Dr. Fernando Koch, and Dr. Alan Kunz-Lomelin |

Enhancing Child Welfare Research and Translation through AI

Keywords: child welfare, foster care, artificial intelligence, generative AI, natural language processing, case work, ethical AI, social impact, social work

Our Research

Child welfare agencies face complex challenges with fragmented records, administrative overload, and incomplete case information. These issues place a heavy burden on practitioners and can affect the quality of decisions that shape children’s lives.
Our research explores how to apply generative intelligence to improve this process. We are building a multidisciplinary platform where social work and engineering students collaborate to design and deploy an AI assistant for child welfare analytics. The solution will take unstructured information from case notes, reports, and other records and turn it into clear insights that are easy to understand and act on. Practitioners will receive concise information and recommendations for key actions, helping them work more effectively, accurately, and with greater confidence.


Our Strategy

Students will gain hands-on experience at the intersection of social impact and advanced AI. You’ll learn how to:

  • Analyze literature and map challenges in current child welfare practice
  • Develop generative AI pipelines for information extraction and summary generation
  • Build risk stratification and prioritization models that highlight high-risk cases
  • Create explainability dashboards that translate predictions into transparent, user-friendly narratives
  • Address challenges in bias mitigation, ethical AI, and responsible deployment

You will also work on co-developing preprocessing tools, natural language processing modules, and predictive models using de-identified datasets. Each prototype will be tested against real-world workflows and the ethical needs of children, families, and professionals.


Our Impact

This project aims to connect cutting-edge AI with frontline child welfare in order to:

  • Streamline data workflows and reduce administrative burden
  • Augment professional judgment with transparent, interpretable insights
  • Help practitioners prioritize at-risk children and families more effectively
  • Mitigate bias and promote equity in child welfare decisions
  • Reduce burnout for frontline workers while improving outcomes for children and families

Our vision is a human-centered, explainable AI framework that makes the child welfare system more accurate, equitable, and sustainable.


Our Team

Dr. Morgan Cooley, Sandler School of Social Work

Dr. Fernando Koch, Department of Electrical Engineering and Computer Science

Dr. Alan Kunz-Lomelin, Sandler School of Social Work