Abstract

Public sector agencies are rapidly adopting AI systems to make critical decisions that span across services such as benefits eligibility, housing and homelessness services, child welfare services, mobility, inspections, and more. This foregrounds long‑standing HCI concerns about participatory design in settings marked by institutional constraints, public accountability, and uneven power dynamics. While there is broad agreement that community‑centered practices are necessary, practitioners still wrestle with three practical questions: (1) which participatory methods fit distinct stages of the AI lifecycle; (2) what to measure to show that engagement influenced problem framing, data, model, design, or policy; and (3) how procurement and vendor management can make participation feasible and verifiable when agencies acquire AI tools rather than build in-house. This workshop will focus on two themes: methods and measures, to identify practical ways of showing how participation shapes design and policy, and procurement and contracts, to explore how engagement requirements can be embedded in vendor agreements.

Document Type

Conference Proceeding

Publication Date

4-13-2026

Publication Title

Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems

First Page

1

Last Page

6

Download

Share

COinS