Decision-Making Frameworks for AI-Enabled Cyber Risk Assessments in Financial Institutions

The perpetually increasing crop of cyber threats forces financial institutions to improve their capacity for measuring and managing cyber risks. This research proposes a structured AI-based method of cyber risk assessment that is tailored to meet the particular needs of the financial institutions. Through use of probabilistic risk frameworks, current threat feeds, and decision-making tools that measure multiple criteria, the framework is developed. Examining the practices of a Tier-1 bank, we show how the framework increases the accuracy of 0.96 % using the machine learning classifiers. Our framework consists of a 5-layer system for data collection, risk assessment, impact evaluation, and mitigation suggestions in addition to immediate decision guidance based on AI and Bayesian networks. After increases in performance observed upon testing the framework against simulated threat vectors, we get to know the proposed approach improvement. Within this research, we demonstrate how AI-enriched frameworks enable flexible and data-driven governance strategies that can successfully adjust to the current transformation of the regulatory requirements as well as the operational needs in finance.