GENERATIVE AI ADOPTION: AN EXPLORATORY INNOVATIVE TOOL FOR CURBING ACCOUNTING FRAUDS IN THE NIGERIAN BANKING SECTOR DATA LAKE ANALYTICS
DOI:
https://doi.org/10.57233/ijamer.v2i3.022Keywords:
Generative Artificial Intelligence, Accounting Fraud, Nigerian Banking Sector, Data Lake Analytics, Synthetic Data, Fraud Detection, Anomaly DetectionAbstract
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
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Author(s)

This work is licensed under a Creative Commons Attribution 4.0 International License.
Google Scholar
CORE
BASE (Bielefeld Academic Search Engine)