Financial fraud continues to pose a significant and persistent challenge for Pakistan’s banking sector, particularly amid rapid digitalization and the expansion of online financial services. Artificial intelligence (AI) has emerged as a critical technological response to these evolving risks. This study examines how AI-based fraud detection and prevention systems are implemented in Pakistani banks and evaluates the extent to which strategic objectives align with operational practices. Adopting an exploratory qualitative design, the study integrates a systematic literature review with in-depth interviews involving branch managers, fraud officers, IT specialists, senior executives, and customers across multiple banking institutions. The findings reveal limited awareness and understanding of AI-driven fraud management among branch-level staff and customers, with responsibility for fraud monitoring remaining highly centralized among senior management. Though advanced techniques such as machine learning, anomaly detection, deep learning, and natural language processing are technically available, their operational utilization remains constrained. Grounded in the Technology Acceptance Model (TAM) and the Resource-Based View (RBV), the current study demonstrates that user acceptance, workforce capability, and organizational readiness critically shape AI effectiveness. The study highlights the urgent need for enhanced FinTech literacy, decentralized AI deployment, and stronger governance mechanisms to bridge the strategic-operational divide. These measures are essential for aligning operational practices with strategic intent and for unlocking AI’s full potential to enhance fraud resilience, regulatory compliance, and customer trust.
Rahim et al. (2026) studied this question.