1-20-2026 8:29am

Public Agenda: Insights and Analysis on Public Service and the People When Algorithms Govern: Racial Equity, Due Process, and Who Answers When AI Harms?

John Williams Editor

Photo by Conny Schneider on Unsplash

by Dr Shellie M Bowman Sr

Public administration has always relied on tools. The difference today is that some tools do not merely support decisions. They shape them. Algorithms and artificial intelligence systems increasingly help governments rank risk, flag fraud, target enforcement, and influence sentencing. In practice, that can mean an automated output becomes the basis for authoritative state action, including denial of benefits, wage garnishment, detention, or harsher punishment.

For communities of color, the stakes are especially high because many high-impact public systems already carry documented inequities in policing, housing, health access, and economic security. An algorithm does not need to explicitly use race to reproduce racialized outcomes. Bias can enter through historical data, unequal access to services, and proxy variables that reflect segregation and structural disadvantage (Barocas & Selbst, 2016; Obermeyer et al., 2019). When the government relies on these systems without safeguards, inequity can be automated, scaled, and defended as objective.

Below are four real-world use cases illustrating when the government relied on artificial intelligence or algorithmic systems to decide, adjudicate, and conclude in ways that established the basis for authoritative execution. Each raises a central question of public administration: who is accountable when the outcome is wrong and damaging?

If you want to read the complete article click on

Any comments place them below or on the Public Agenda page on Substack
https://open.substack.com/pub/drshellieb/p/when-checks-and-balances-str…

Stay Connected with Public Agenda by Dr. Shellie M. Bowman