Governing Civic Algorithms: A Procedural Justice Framework for AI-Mediated Workforce Systems

• In Event: Participation and Procedural Justice in AI Governance

Abstract

Public workforce agencies increasingly rely on algorithmic systems to triage services, match job seekers to opportunities, and detect fraud. While these systems promise efficiency, they also introduce new governance challenges, particularly when algorithmic decisions shape access to public benefits and services without adequate transparency, explanation, or recourse.

This paper introduces the concept of civic algorithms—algorithmic systems that function as public institutions by mediating access to government opportunities, benefits, and services. We argue that when these systems operate without appropriate procedural safeguards, they create “procedural gaps” that increase administrative burden, reduce program engagement, and amplify inequities.

Drawing on procedural justice, administrative law, and sociotechnical governance theory, we propose the Civic Algorithm Governance Standard (CAGS), a five-step framework that establishes minimum requirements for transparency, human oversight, data governance, equity monitoring, and public accountability. The framework is designed as an artifact-based governance model, enabling evaluation and oversight under real administrative conditions without requiring access to proprietary systems or individual-level data.

We demonstrate the diagnostic utility of CAGS through a retrospective application to Michigan’s MiDAS unemployment fraud detection system, a widely documented case of large-scale algorithmic failure. This application illustrates how gaps in procedural governance can scale into systemic harm, and how structured governance standards could support earlier detection and mitigation.

From a policy perspective, this paper contributes to ongoing efforts to operationalize responsible AI in public systems by translating high-level principles into actionable governance requirements. It also highlights the need for workforce and digital governance frameworks that treat algorithmic systems not merely as technical tools, but as institutional actors subject to standards of procedural legitimacy and public accountability.

References

Eubanks, Virginia. 2018. Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor. New York: St. Martin’s Press.

Moynihan, Donald P., Pamela Herd, and Hope Harvey. 2015. “Administrative Burden: Learning, Psychological, and Compliance Costs in Citizen-State Interactions.” Journal of Public Administration Research and Theory 25(1): 43–69.

Author

• DORCAS U ATTAH, Workforce Intelligence Institute
  Presenting Author