GENERATIVE AI ADOPTION: AN EXPLORATORY INNOVATIVE TOOL FOR CURBING ACCOUNTING FRAUDS IN THE NIGERIAN BANKING SECTOR DATA LAKE ANALYTICS

Authors

  • Emoarehi Eriki PhD Department of Accounting, Faculty of Management Science, University of Benin, Benin City Author
  • Samuel Umanah PhD Department of Accounting Faculty of Management Science University of Benin, Benin City. Author

DOI:

https://doi.org/10.57233/ijamer.v2i3.022

Keywords:

Generative Artificial Intelligence, Accounting Fraud, Nigerian Banking Sector, Data Lake Analytics, Synthetic Data, Fraud Detection, Anomaly Detection

Abstract

Accounting fraud remains a significant concern within the Nigerian banking sector, particularly as banking transactions become increasingly complex, digital and data-intensive. Traditional fraud detection approaches may face limitations in identifying variations of fraudulent behaviour that are not adequately represented in historical transaction records. This conceptual paper explores Generative Artificial Intelligence (Gen AI) as an innovative analytical tool for strengthening accounting fraud detection within Nigerian banking sector data lake environments. The paper adopts a conceptual and exploratory approach based on the synthesis of existing literature on Generative AI, accounting fraud detection, synthetic data generation, anomaly detection and banking data analytics. A conceptual framework is proposed linking banking data lakes, data preparation and governance, Generative AI applications, synthetic fraud data and fraud scenario generation, anomaly detection, human validation and accounting fraud prevention. In the framework, Generative Adversarial Networks (GANs) are proposed for generating synthetic fraud-related transaction patterns, while Large Language Models (LLMs) are proposed for generating and interpreting fraud scenarios. The paper argues conceptually that these capabilities may complement existing fraud detection systems by expanding the range of fraud scenarios available for analysis. However, the study recognizes important concerns relating to data privacy, bias, model reliability, security, regulatory compliance and the potential for inaccurate or unrealistic AI-generated outputs. Accordingly, Generative AI is positioned as a complementary tool requiring human oversight rather than a replacement for professional judgement. The paper concludes by recommending empirical pilot studies, regulatory collaboration, capacity building and appropriate AI governance to determine the practical effectiveness and responsible application of Generative AI in Nigerian banking fraud detection.

Downloads

Download data is not yet available.

Author Biographies

  • Emoarehi Eriki PhD, Department of Accounting, Faculty of Management Science, University of Benin, Benin City

    Department of Accounting,

    Faculty of Management Science,

    University of Benin, Benin City

  • Samuel Umanah PhD, Department of Accounting Faculty of Management Science University of Benin, Benin City.

    Department of Accounting

    Faculty of Management Science

    University of Benin, Benin City.

Downloads

Published

2026-09-25

How to Cite

GENERATIVE AI ADOPTION: AN EXPLORATORY INNOVATIVE TOOL FOR CURBING ACCOUNTING FRAUDS IN THE NIGERIAN BANKING SECTOR DATA LAKE ANALYTICS. (2026). International Journal of Accounting, Management and Economic Review, 2(3), 400-414. https://doi.org/10.57233/ijamer.v2i3.022

Similar Articles

11-20 of 210

You may also start an advanced similarity search for this article.