• Version
• Download 1
• File Size 673.97 KB
• File Count 1
• Create Date 8 December 2024
• Last Updated 8 December 2024

Integrating AI in Identity and Access Management for Improved Cybersecurity Posture

Ranga Premsai, MS, IAM Professional, USA

Abstract—The rise of cybercrime has made fraud detection a critical focus for financial services, where even minor improvements in detection rates can translate into significant savings. Traditional rule-based detection systems are increasingly unable to keep up with the complexity and evolving nature of fraudulent schemes. This work explores the role of artificial intelligence (AI), particularly deep learning, in enhancing fraud detection capabilities. Here we propose a novel AI-driven fraud detection system that incorporates Single Sign-On (SSO) identity and access management (IAM) frameworks, leveraging a dual-layered approach with batch and real-time processing.

The batch layer establishes a user trust identity by analysing historical behavioural patterns, which informs an access-granting mechanism that evaluates real-time transaction data for fraud indicators. A deep iterative convoluted memory classifier then assesses transactions, for detecting frauds. This architecture automates fraud detection, allowing analysts to focus on high-priority cases and ensuring that only authenticated users can access sensitive information. Our experiments are Conducted in the MATLAB environment, From the analysis it was revealed that the model's effectiveness, showing promise for scalable, efficient fraud prevention in financial services. This work underscores the transformative potential of AI in securing financial transactions against sophisticated cyber threats.

Index Terms— cybercrime, Fraud, Financial Transactions, artificial intelligence, identity and access management, fraud detection,