NEXT-GENERATION AI-ENABLED BIOMETRICS FOR FRAUD PREVENTION, DIGITAL BANKING, AND FINANCIAL SECURITY USING IRIS RECOGNITION

Abstract

Background: The high growth in digital banking has led to more demand in the use of enhanced, secure, and user-friendly authentication. Iris recognition is a next-generation biometric technology that is highly accurate, does not spoof, and is more convenient to use. AI-enabled iris recognition can enhance biometric authentication by applying intelligent pattern recognition to iris features, supporting more accurate identity verification and stronger protection against fraudulent access in digital banking environments. Nonetheless, there is little empirical evidence regarding its acceptance and perceived security and influence on digital banking security. These AI-enabled biometric capabilities are particularly relevant to fraud prevention and financial security because they can strengthen identity verification and reduce opportunities for unauthorized access to digital banking services.
Objectives: This paper explores the effectiveness of technological features of iris recognition in digital banking, such as accuracy, anti-spoofing, system response time, and user interface experience in influencing the security of digital banking as well as user adoption. It also looks into the mediating role of the perceived security as well as the moderating role of trust in biometric technologies. The study also considers the relevance of AI-enabled iris recognition to fraud prevention and financial security by examining whether enhanced biometric authentication can strengthen digital banking security and reduce the risk of unauthorized access.
Methods: The quantitative research design was used, and the data were gathered using 233 respondents (digital banking users) through a structured Likert-scale questionnaire. The Shapiro-Wilk, Cronbach's Alpha, KMO, and Bartlett tests were used to confirm normality, reliability, as well as validity. Independent samples t-test, One-way ANOVA, Chi-square, and Kruskal-Wallis tests were used to analyze group differences. Pearson correlation and multiple regression were the two methods employed to test the relationship between variables. The questionnaire therefore assessed the perceived effectiveness of AI-enabled iris recognition, including biometric accuracy, anti-spoofing capability, security, and authentication performance in relation to fraud prevention and financial security.
Results: The reliability of all the instruments was excellent (α ≥ 0.70), and the validity (KMO = 0.842). Results showed there were strong and positive relationships between all variables. Findings of the regression tests proved the existence of a significant correlation between accuracy, anti-spoofing technology, system response time, user experience, perceived security, and trust in increasing user adoption of the iris recognition. The strongest predictor was perceived security. The difference tests of the group presented the results that gender, age, education level, and biometric experience had a significant effect on security perceptions and adoption intentions. The findings further indicate that AI-enabled iris recognition capabilities, particularly accuracy, anti-spoofing performance, and perceived security, are positively associated with stronger digital banking protection and fraud-prevention potential.
Conclusion: The paper has come to the conclusion that iris recognition is a highly potent and efficient biometric remedy to improving the security of digital banking. High perceived security and trust in conjunction with technological performance create a great user adoption. With the application of awareness and trust-building techniques and user-centric design, iris recognition has high potential of becoming a mainstream authentication in financial services. The findings further support the potential of AI-enabled iris recognition as an intelligent biometric security mechanism for fraud prevention and strengthening financial security in digital banking.