From Compliance to Intelligence: Integrating AI and Predictive Analytics into U.S. Tax Compliance and Revenue Systems

Abstract

The gross tax gap in the U.S is over 600 billion a year. AI and predictive analytics have a transformational opportunity to enhance compliance risk scoring, audit selection, revenue forecasting, and fraud detection at the IRS. This critical literature review summarizes peer-reviewed literature (20202026) on AI adoption to the U.S. tax compliance and revenue systems, evaluating reported outcomes, methodological conflicts, and governance needs. Organized search in SSRN, Google Scholar, Web of Science, Scopus, and government repositories resulted in 37 sources that satisfy pre-determined inclusion criteria and are rated in three levels of evidence. The ML models decrease audit no-change rates by an estimated 15-20 percentage points and forecasting MAPE by 15-30 percent compared to legacy systems, although equity questions are actualized: ROI-optimal classifiers increase audit load on low-income filers unless fixed through regression-based expected-adjustment targets and fairness limits. Governance alignment with NIST AI RMF 1.0, EO 14110, and 26 U.S.C. § 6103 is critical. Equity-by-design, explainable architecture, federated infrastructure, and modernized statutory framework are all necessary to achieve responsible AI adoption.