A Systematic Review of AI-Driven Financial Information Systems for Budget Optimization in U.S. Public Sector Institutions (2018–2026)

Abstract

This study examines the problem that many U.S. public sector institutions collect large volumes of financial and operational data but still face difficulties converting those data into accurate budget forecasts, efficient allocation decisions, real-time expenditure monitoring, fraud detection, and accountable financial governance. The purpose of the study is to assess how AI-driven financial information systems support budget optimization in public institutions through predictive analytics, machine learning, robotic process automation, anomaly detection, intelligent dashboards, and decision-support algorithms. The research follows a quantitative, cross-sectional, case-based design supported by systematic evidence synthesis from 35 selected studies and documented U.S. public sector cases, including federal agencies, state departments, local governments, public universities, public healthcare institutions, public transportation agencies, and public procurement offices. The key variables include AI-driven financial information systems, forecasting accuracy, budget allocation efficiency, expenditure monitoring, anomaly detection, transparency, accountability, data quality, institutional readiness, cybersecurity maturity, and AI governance. The analysis plan involved structured screening, coding, thematic synthesis, descriptive frequency analysis, percentage distribution, and five-point Likert-style evidence scoring to evaluate the strength of literature-based findings. The headline findings show that 29 of 35 studies, or 82.9%, reported that AI improved at least one major budget function, with an overall evidence score of 4.25 out of 5.00. Forecasting accuracy was supported by 27 studies, or 77.1%, with a score of 4.32; allocation efficiency was supported by 24 studies, or 68.6%, with a score of 4.08; expenditure monitoring and fiscal accountability were supported by 26 studies, or 74.3%, with a score of 4.21; and institutional readiness and governance factors were identified in 28 studies, or 80.0%, with the highest score of 4.40. The findings imply that public agencies should strengthen data quality, system integration, staff capability, cybersecurity safeguards, explainable AI governance, and human oversight to ensure that AI-driven financial information systems improve budget credibility, reduce waste, enhance fraud control, and support transparent public value creation.