The Growing Role of AI in U.S. Government Decisions

California has recently acknowledged its use of several high-risk artificial intelligence (AI) systems. This revelation underscores a significant development: AI is already influencing U.S. government decisions impacting millions of citizens.

According to a report by Cal Matters in June 2026, California officials revealed the existence of six high-risk AI systems used for tasks ranging from fraud detection to education oversight. Previously, the state had reported no such systems. These tools identify potential unemployment fraud, oversee university examinations, and detect AI-generated student work.

This incident has raised concerns about transparency. Experts worry that if governments cannot effectively monitor their own AI systems, the public may lack awareness of how automated tools affect crucial decisions. This concern is vital as AI has moved beyond experimental stages. Across the U.S., states are deploying algorithms for criminal justice, benefits processing, student monitoring, and transportation management, directly impacting rights, finances, and daily routines.

AI in Criminal Justice

AI’s application in criminal justice is particularly contentious. In California, AI predicts whether incarcerated individuals will reoffend. States like Pennsylvania have developed tools to assist judicial decision-making.

Systems like COMPAS operate in states such as Pennsylvania, New York, Wisconsin, and Florida. They generate risk scores influencing bail, parole, and sentencing decisions. Critics question whether these systems perpetuate biases. Supporters argue they offer consistent, data-driven insights.

AI in Benefits Decisions

AI’s role in benefits management is growing. States use it to identify fraud and handle claims. California employs automated systems for unemployment claims, while Nevada uses AI to draft unemployment appeal rulings for human review.

Automation risks have surfaced. Michigan’s unemployment fraud detection system erroneously labeled thousands as fraudsters, prompting restored human oversight—highlighting the need for stricter safeguards.

AI in Education

Education systems are increasingly adopting AI. California utilizes tools to spot AI-generated assignments. Wisconsin uses an algorithm-driven early-warning system predicting students at risk of not graduating.

Schools are experimenting with AI-powered tutoring, personalized learning, and administrative tools, reflecting the expanding role of technology in classrooms nationwide.

AI in Transportation

Transportation departments leverage AI for congestion management, infrastructure monitoring, and real-time operational decisions. States like Texas, Tennessee, and North Carolina utilize systems analyzing traffic data to optimize roadway operations.

Proponents believe these systems enhance efficiency and reduce congestion. Critics voice concerns regarding oversight and accountability.

State Leaders in AI Implementation

While AI adoption is widespread, certain states lead the way. According to GL Solutions, these include:

• California: Utilizes AI across criminal justice, benefits, and government services.
• Texas: Employs AI in Medicaid oversight and fraud detection.
• New York: Deploys AI-driven systems for unemployment fraud detection and consumer protection.
• Virginia: Implements AI-powered traffic management.

Other states such as Colorado, Utah, and Virginia are also developing AI policy frameworks or pilot programs, though specifics vary.

The Nationwide AI Shift

AI is no longer experimental within state governments. It integrates deeply into systems dictating public service delivery and agency decision-making.

Almost every state has explored or implemented AI, spanning fraud detection, education, transportation, and administration. Adoption accelerates as agencies face staffing shortages, increased workloads, and demands for efficiency.

The debate is shifting from whether governments should use AI to how these systems should be governed. As AI embeds more in government operations, stakeholders question the levels of transparency and accountability necessary when algorithms influence vital decisions.