AI-Enabled Audit Analytics for SME Financial Reporting and Anomaly Detection: A Risk-Based Framework for Early irregularity identification and Control Strengthening

Small and Medium Enterprises (SMEs) are susceptible to a "governance gap" due to weak controls and a high risk of financial fraud. This paper presents a Risk-Based Audit Analytics Framework to address this gap by combining AI-powered anomaly detection with transaction pattern analysis. By drawing on Shem & Mupa's (2024) research on business rescue and stakeholder engagement, and current research on the application of AI in fraud prevention (Adeboye, 2024; Antwi et al., 2024), the research explores how SMEs can use data science to identify unusual activity sooner than traditional audit processes.
We examine the use of Machine Learning (ML) in general ledger anomaly detection and Process Mining to assess internal controls (Duan et al., 2024). This framework builds on Mupa's "Enterprise Resilience" by suggesting that, to survive financial hardship, an SME must adopt a Sustainable Auditing model that offers real-time transparency (Pillai, 2025). By analyzing the use of AI in financial reporting and the ethical considerations of automated auditing (Murikah et al.2024), this study offers a scalable approach to enhancing controls in resource-limited settings. The article concludes that AI-powered analytics not only enhance efficiency but serve as a "digital buffer" for SME survival and community prosperity (Ayankoya et al., 2025).