AI-powered fraud detection in digital banking: Enhancing security through machine learning

This study examines the role of Artificial Intelligence (AI) and Machine Learning (ML) in enhancing fraud detection within the digital banking sector. With financial transactions migrating to the digital platforms, sophistication of the fraudsters comes in and advanced security measures are needed. Machine Learning models that support the AI driven fraud detection systems, analyse huge datasets, find the anomalies and reduce the risk of financial fraud. In this literature review, this author critically evaluates existing AI/ML based fraud detection methods in terms of the effectiveness of the methods, the challenges faced by the methods, and avenues of what is scaled up more towards them being a solution. The review identifies key trends on supervised and unsupervised learning, deep learning models, and the findings on the anomaly detection technique. The findings highlight AI’s capacity for enhancing the accuracy of fraud detection whilst tackling algorithmic bias, the privacy of data and the attack of adversarial. The study ends by providing recommendations for enhancing the fraud detection system in terms of the use of Explainable AI (XAI), real time fraud monitoring, and integrating blockchain into digital banking security.