Purpose: The aim of this study is to identify and prioritize artificial intelligence (AI) applications in the banking industry and to develop a decision-making framework to assist bank managers in allocating resources and selecting investment priorities for AI projects.



Methodology: In terms of its objective, this study is applied, while methodologically it employs a descriptive-analytical design based on a multi-criteria decision-making approach. First, a structured review of previous studies will be conducted to identify the major applications of AI and the criteria influencing their prioritization. Subsequently, the preliminary criteria and alternatives will be validated and refined using the Delphi method and the opinions of experts in banking, information technology, and artificial intelligence. After determining the final criteria, the Best-Worst Method (BWM) will be used to determine the weights of the criteria, while the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) will be employed to rank AI applications.



Findings: Based on the review of previous studies, applications such as fraud detection, credit assessment and credit scoring, risk management, anti-money laundering (AML), customer services and chatbots, process automation, identity verification, and cybersecurity have been identified as preliminary alternatives. In addition, six preliminary dimensions have been extracted for evaluating and prioritizing these applications: economic value, operational value, risk and security value, customer value, implementation feasibility, and regulatory requirements and responsible AI.



Originality/ Value: The added value of this study lies in developing a dedicated framework for the banking industry. Unlike studies that focus solely on AI applications or their adoption factors, the proposed framework simultaneously compares multiple AI applications based on a comprehensive set of economic, operational, risk and security, customer, implementation feasibility, regulatory, and responsible AI criteria. The integration of a literature review, Delphi method, BWM, and TOPSIS enables the combination of scientific evidence and expert judgment and provides a systematic decision-making tool for prioritizing AI investments in the banking sector.