Artificial Intelligence–Driven Audit Analytics and Fraud Detection Effectiveness in Public Sector Procurement Systems: A Quantitative Study in Nigeria

Abstract

Public sector procurement systems remain highly vulnerable to fraud, inefficiency, and weak accountability structures, particularly in developing economies where manual audit processes dominate financial oversight. This study examines the effect of artificial intelligence (AI)–driven audit analytics on fraud detection effectiveness in public sector procurement systems using a quantitative IMRaD structure aligned with Scopus-indexed journal standards. The study adopted a descriptive survey research design with a population of 1,000 respondents drawn from procurement officers, internal auditors, ICT personnel, and compliance units in selected public institutions in Nigeria. A sample size of 286 respondents was determined using the Taro Yamane sampling formula. Data were collected through structured questionnaires and analysed using descriptive statistics, Pearson correlation, and multiple regression analysis. Findings reveal that AI-driven anomaly detection, predictive analytics, machine learning-based risk scoring, and real-time procurement monitoring significantly enhance fraud detection effectiveness. However, inadequate digital infrastructure, resistance to technological change, and limited AI competencies constrain full system optimisation. Regression results indicate a strong and statistically significant relationship between AI-driven audit analytics and fraud detection effectiveness. The study concludes that artificial intelligence significantly strengthens procurement transparency and audit efficiency while reducing fraud exposure in public sector systems. It recommends accelerated digital transformation, institutional capacity building, and full integration of AI audit tools in procurement governance frameworks.