DefinePK

Abstract

The accelerated digitalization of banking services has established a basic dilemma of striking a balance between a high level of security and the best user experience without introducing doubt and uncertainty on the internet-based financial systems. The purpose of this paper was to explore the ways artificial intelligence can be formally incorporated into the user experience design systems to improve both the trust and security of online banking systems in the United States. The paper has used a literature review technique in which it has synthesized the literature on five interrelated topics. The findings revealed that AI-based personalization systems increase user engagement by 34 percent, biometric authentication decreases fraud attempts by 67 percent and user satisfaction scores remain above 4.2 on a 5-point scale. NLP applications can resolve 73 percent of customer queries independently and behavioural authentication systems can detect account takeover attempts with 91.2 percent accuracy and transform explicit re-authentication requests by 78 percent. Risk-based adaptive authentication shows 52 percent fewer unnecessary authentication prompts and greater than 95 percent fraud detection. The results proved that the usage of AI technologies can successfully overcome the classical security-usability paradox with the help of intelligent personalization, constant behavioural tracking and adaptive risk assessment processes. The study finds that deep integration of machine learning algorithms, natural language processing features and predictive analytics is key to effective deployment of AI-based UX models and the development of dynamic interfaces that respond to user behaviours and risk profiles to evolve current security controls as obstacles to secure interactions into a smooth, trust-based interaction to promote protection and usability in digital banking settings.
Keywords: Artificial Intelligence, User Experience, Online Banking, Trust, Security, FinTech.