A Scalable Enterprise Architecture for Secure and Autonomous Generative AI Systems in Banking

Abstract

Generative AI (GenAI) has a strong potential to transform the business of banking by automating and improving the effectiveness of decision-making. Yet, its incorporation in the regulated financial settings poses security, compliance and transparency related challenges. This paper presents a suggested scalable enterprise architecture to facilitate the implementation of secure and autonomous GenAI systems in the banking industry that would be in compliance with the regulatory and protect the sensitive financial information stored. The architecture also includes approval processes, audit functionality, and security to achieve high standards of regulation. The study will also determine the performance with respect to the effectiveness of this architecture to enable the integration of GenAI and the response to compliance, security, and scalability issues. The results can be used to build safe AI implementation models in the banking industry.