Module 6: Systemic Control: Race, AI, and the Justice System

Systemic Control: Race, AI, and the Justice System

Introduction: The Architecture of Inequality

We now turn our attention to the institution that represents the state's ultimate power of social control: the criminal justice system. A critical sociological analysis reveals that this system is not a neutral arbiter of law but an active agent in the maintenance of social hierarchy.^[1] The central thesis of this module is that the social constructions of race and crime have historically converged to create a system of profound racial disparity, and that emerging Artificial Intelligence (AI) technologies threaten to automate and amplify these historical injustices under a veneer of computational neutrality.^{[2, 3]}

Part 1: The Dual Constructions of Social Control

The Social Construction of Race and Crime

To comprehend the mechanics of systemic racism within the justice system, we must deconstruct its foundational concepts. Race is not a biological reality but a social construct—a "badge of difference" used to justify unequal treatment.^{[5, 6]} Similarly, crime is also socially constructed. An act becomes criminal only when a society's dominant group defines it as such in law, often criminalizing behaviors associated with marginalized groups while minimizing harms perpetrated by the powerful.^{[7, 8]}

"If men define situations as real, they are real in their consequences." - W.I. Thomas^[6]

This process has forged a deep, socially reinforced association between blackness and criminality in the American consciousness, influencing decision-making at every level of the justice system.^[8] The result is the criminalization not of an act, but of an identity.

The Statistical Reality of Disparity

The tangible consequences are visible in stark racial disparities. Black individuals are incarcerated in state prisons at nearly five times the rate of whites.^[17] For Black men born in 2001, the lifetime likelihood of imprisonment is a shocking one in five.^{[13, 19]} These disparities are not isolated incidents but are cumulative, creating an "avalanche" effect where small biases at each stage of the system compound to produce staggering levels of overrepresentation.^[20]

Part 2: The New Frontier: AI in the Justice System

Proponents champion AI as a solution to human bias, but a critical examination reveals a more troubling reality. Trained on historical data from a biased system, AI risks inheriting, amplifying, and concealing injustice under a deceptive cloak of objectivity—a process known as "bias laundering."^{[23, 24, 25]}

Predictive Policing: These systems analyze historical arrest data to predict crime "hotspots," creating a pernicious feedback loop. Biased data leads to targeted deployment in minority communities, which leads to more arrests, which is then fed back as "confirmation" to the algorithm, amplifying the cycle.^{[25, 27, 28]}
Risk Assessment Algorithms: Tools like COMPAS generate "risk scores" to predict recidivism. A landmark ProPublica investigation found these tools were racially biased, falsely flagging Black defendants as high-risk at nearly twice the rate of white defendants.^{[32, 33, 34]}
Facial Recognition: This technology is demonstrably less accurate for people of color and women. Every publicly reported case of a wrongful arrest based on a false facial recognition match has involved a Black person.^{[41, 42, 44]}

Part 3: Applying Sociological Theory

The major sociological paradigms provide powerful lenses to explain these dynamics.

Conflict Theory / Critical Race Theory views the justice system as an instrument of domination, designed by the powerful to control subordinate populations.^{[47, 48]} AI is seen not as a neutral tool, but as a new, highly efficient weapon in the arsenal of social control.^[51]
Structural Functionalism analyzes the system's functions. Its manifest function is maintaining social order.^[9] However, mass incarceration and the erosion of legitimacy are severe latent dysfunctions that threaten the stability of the entire society.^{[12, 20]}
Symbolic Interactionism / Labeling Theory focuses on how the label "criminal" is disproportionately applied to people of color, becoming a master status or stigma.^{[54, 55]} An AI-generated "risk score" becomes a powerful and seemingly objective digital stigma—a "super-label" that can create a self-fulfilling prophecy.^{[25, 57]}

Part 4: Deconstructing the "Law and Order" Narrative

The operation of the justice system is shaped by political rhetoric. The "law and order" narrative, popularized in the 1960s, frames crime in individualistic and moralistic terms, attributing it to "bad people" rather than social conditions.^[58] This rhetoric justifies punitive policies that disproportionately harm communities of color by framing the symptoms of inequality as its cause.^{[59, 60]} Competing narratives focused on public health, community investment, and restorative justice offer alternative frames that seek to address the root causes of crime rather than simply punishing its outcomes.^{[12, 61, 62]}

Visualizing Systemic Control

Key data points that reveal the staggering scale of racial disparity in the U.S. criminal justice system.

Incarceration Disparity

Black Americans are incarcerated in state prisons at nearly five times the rate of white Americans.^[17]

1 in 5

Lifetime Risk (Black Men)

is the lifetime likelihood of imprisonment for a Black man born in 2001, a rate four times that of their white counterparts.^{[13, 19]}

Algorithmic Bias

The COMPAS algorithm was nearly twice as likely to falsely label Black defendants as "high-risk" compared to white defendants.^{[32, 34]}

The Predictive Policing Feedback Loop

1. Biased Data Input

Algorithm is trained on historical data reflecting over-policing in minority neighborhoods.

➞

2. Targeted Deployment

AI directs more police patrols into these same neighborhoods, labeling them "hotspots."

➞

3. Increased Arrests

Heavier police presence naturally leads to more arrests for minor offenses in these areas.

➞

4. Biased "Confirmation"

New arrest data is fed back into the AI, which sees it as "proof" its prediction was correct, amplifying the cycle.

This chart illustrates the stark disparity in state prison incarceration rates per 100,000 people by race and ethnicity.^[17]

The Ecology of Systemic Control

An ecological model showing how macro-level racism is enacted through institutions and experienced by individuals. Hover over a node or level to learn more.

References

Social Control. (n.d.). American Sociological Association.
Systemic Racism. (n.d.). The Aspen Institute.
Alexander, M. (2010). *The New Jim Crow: Mass Incarceration in the Age of Colorblindness*.
Muhammad, K. G. (2010). *The Condemnation of Blackness: Race, Crime, and the Making of Modern Urban America*.
Du Bois, W. E. B. (1903). *The Souls of Black Folk*.
Thomas, W. I., & Thomas, D. S. (1928). *The Child in America: Behavior Problems and Programs*.
Becker, H. S. (1963). *Outsiders: Studies in the Sociology of Deviance*.
Eberhardt, J. L., et al. (2004). Seeing Black: Race, Crime, and Visual Processing. *Journal of Personality and Social Psychology*.
Reiman, J., & Leighton, P. (2017). *The Rich Get Richer and the Poor Get Prison*.
ACLU. (2020). *A Tale of Two Countries: Racially Targeted Arrests in the Era of Marijuana Reform*.
Pierson, E., et al. (2020). A large-scale analysis of racial disparities in police stops across the United States. *Nature Human Behaviour*.
The Sentencing Project. (2023). *One in Five: Disparities in Crime and Policing*.
The Sentencing Project. (2023). *One in Five: Ending Racial Inequity in Incarceration*.
SAMHSA. (2021). *National Survey on Drug Use and Health*.
Bureau of Justice Statistics. (2021). *Arrests*.
U.S. Sentencing Commission. (2023). *2023 Demographic Differences in Federal Sentencing*.
Nellis, A. (2021). *The Color of Justice: Racial and Ethnic Disparity in State Prisons*. The Sentencing Project.
The Sentencing Project. (n.d.). State-by-State Data.
Bureau of Justice Statistics. (2022). *Jail Inmates in 2021*.
Pager, D. (2007). *Marked: Race, Crime, and Finding Work in an Era of Mass Incarceration*.
National Institute of Justice. (n.d.). *Artificial Intelligence (AI)*.
RAND Corporation. (2020). *The Promise and Perils of Artificial Intelligence and the Criminal Justice System*.
O'Neil, C. (2016). *Weapons of Math Destruction*.
Eubanks, V. (2018). *Automating Inequality*.
Noble, S. U. (2018). *Algorithms of Oppression*.
Upturn. (2016). *Predictive Policing Explained*.
Lum, K., & Isaac, W. (2016). To predict and serve? The limits of police computation. *Big Data & Society*.
Brayne, S. (2020). *Predict and Surveil: Data, Discretion, and the Future of Policing*.
O'Donnell, K. (2020). Challenging Racist Predictive Policing Algorithms Under the Equal Protection Clause. *New York University Law Review*.
Ensign, D., et al. (2018). Runaway Feedback Loops in Predictive Policing. *Proceedings of the 1st Conference on Fairness, Accountability and Transparency*.
Richardson, R., Schultz, J., & Crawford, K. (2019). Dirty Data, Bad Predictions. *New York University Law Review*.
Northpointe. (2015). *COMPAS Risk Assessment*. [Now Equivant]
Angwin, J., Larson, J., Mattu, S., & Kirchner, L. (2016). Machine Bias. *ProPublica*.
Dressel, J., & Farid, H. (2018). The accuracy, fairness, and limits of predicting recidivism. *Science Advances*.
AI Now Institute. (2019). *AI in the Criminal Justice System*.
Reisman, D., et al. (2018). Algorithmic Impact Assessments. *AI Now Institute*.
Barocas, S., & Selbst, A. D. (2016). Big Data's Disparate Impact. *California Law Review*.
ACLU. (2023). *How Artificial Intelligence Might Prevent You From Getting Hired*.
Goodman, C. C. (2025). Algorithmic Bias and Accountability. *University of Colorado Law Review*.
Raghavan, M., et al. (2020). Mitigating bias in algorithmic hiring. *Proceedings of the 2020 conference on fairness, accountability, and transparency*.
Buolamwini, J., & Gebru, T. (2018). Gender Shades. *Proceedings of the 1st Conference on Fairness, Accountability and Transparency*.
Hill, K. (2020, June 24). Wrongfully Accused by an Algorithm. *The New York Times*.
Electronic Frontier Foundation. (n.d.). *Facial Recognition*.
Garvie, C., Bedoya, A., & Frankle, J. (2016). *The Perpetual Line-Up*. Georgetown Law Center on Privacy & Technology.
NIST. (2019). *Face Recognition Vendor Test (FRVT) Part 3: Demographic Effects*.
ACLU. (2022). *The Government’s Use of Facial Recognition Technology*.
Macionis, J. J. (2021). *Sociology* (18th ed.).
Lynch, M. J. (2020). *The New Primer in Radical Criminology*.
Bonger, W. (1916). *Criminality and Economic Conditions*.
Quinney, R. (1974). *Critique of Legal Order: Crime Control in Capitalist Society*.
Crenshaw, K., et al. (Eds.). (1995). *Critical Race Theory: The Key Writings That Formed the Movement*.
Durkheim, É. (1893). *The Division of Labor in Society*.
Clear, T. R. (2007). *Imprisoning Communities: How Mass Incarceration Makes Disadvantaged Neighborhoods Worse*.
Goffman, E. (1963). *Stigma: Notes on the Management of Spoiled Identity*.
Becker, H. S. (1963). *Outsiders: Studies in the Sociology of Deviance*.
Lemert, E. M. (1951). *Social Pathology*.
Chiricos, T., & Waldo, G. P. (1975). Socioeconomic Status and Perceived Severity of Crime. *Social Problems*.
Hinton, E. (2016). *From the War on Poverty to the War on Crime*.
Murakawa, N. (2014). *The First Civil Right: How Liberals Built Prison America*.
Gottschalk, M. (2014). *Caught: The Prison State and the Lockdown of American Politics*.
The Justice Collaborative Institute. (n.d.). *Redefining Public Safety*.
Zehr, H. (2002). *The Little Book of Restorative Justice*.