Evaluating AI-Driven tools for improving Digital Literacy and Fraud Risk Detection in Indian Public Sector Banks

Abstract

The development of Artificial Intelligence (AI) in the banking industry is rapidly expanding new possibilities, both in terms of focusing on the lack of digital literacy in banking staff and managing the increasing risks of fraudulent operations during deals. The present research assesses the effectiveness of AI-based tools and their ability to enhance digital literacy skills and increase the potential of fraud risk detection among the employees of Indian state-owned banking institutions. A mixed-methods research design was adopted, which determined the quantitative data among 312 employees in 14 authorities of the public sectors through a structured questionnaire that was validated by Confirmatory Factor Analysis (CFA). Various statistical methods such as Pearson Correlation, Multiple Regression Analysis, One-Way ANOVA, Paired Sample t-test, Kruskal-Wallis H-test, Chi-Square Analysis and Structural Equation Modelling (SEM) are used to test six hypotheses of the study. Findings indicate that the awareness of AI tools ( 0.412, p < 0.001), frequency of employee training ( 0.198, p < 0.001), and implementation readiness ( 0.193, p < 0.001) are important predictors of digital literacy improvements. The effectiveness of detection of fraud was higher after AI intervention (pre-mean = 61.4, post-mean = 78.9; t = -18.654, p < 0.001). The proposed theoretical framework was validated because the SSM model fitted adequately (CFI = 0.947; RMSEA = 0.048). The research also adds to the literature by introducing an empirically tested model that provided an association between the adoption of AI tools and the impact on digital competency and fraud risk reduction outcomes in the Indian banking industry.