The Promises and Perils of using LLMs for Effective Public Services

Authors:
  • EMErina Seh-Young Moon
    Erina Seh-Young Moon
  • MTMatthew Tamura
    Matthew Tamura
  • AZAngelina Zhai
    Angelina Zhai
  • NHNuzaira Habib
    Nuzaira Habib
  • BSBehnaz Shirazi
    Behnaz Shirazi
  • AKAltaf Kassam
    Altaf Kassam
  • DSDevansh Saxena
    Devansh Saxena
  • SGShion Guha
    Shion Guha
Preprint
Publisher website
Google Scholar
Abstract
Governments are the primary providers of essential public services and are responsible for delivering them effectively. In high-stakes decision-making domains such as child welfare (CW), agencies must protect children without unnecessarily prolonging a family's engagement with the system. With growing optimism around AI, governments are pushing for its integration but concerns regarding feasibility and harms remain. Through collaborations with a large Canadian CW agency, we examined how LocalLLM and BERTopic models can track CW case progress. We demonstrate how the tools can potentially assist workers in opportunistically addressing gaps in their work by signaling case progress/deviations. And yet, we also show how they fail to detect case trajectories that require discretionary judgments grounded in social work training, areas where practitioners would actually want support to pre-emptively address substantive case concerns. We also provide a roadmap of future participatory directions to co-design language tools for/with the public sector.
All Related Versions

This publication has 97 references indexed in Scilit: