Artificial Intelligence for stress testing and risk assessment in financial institutions

 Corresponding author: Oyindamola Omolara Ogunruku. Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution License 4.0. Artificial Intelligence for stress testing and risk assessment in financial institutions Oyindamola Omolara Ogunruku * Department of Accounting Finance Economics and Decisions, Western Illinois University, USA. World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 Publication history: Received on 16 May 2025; revised on 23 June 2025; accepted on 25 June 2025 Article DOI: https://doi.org/10.30574/wjarr.2025.26.3.2437 Abstract Banks and financial institutions are using Artificial Intelligence to change how they handle stress testing and risk assessment. This review looks at current AI applications in financial risk management and examines how well these technologies work compared to traditional methods. Traditional financial stress testing faces challenges with nonlinear dependencies and emerging risks, while deep learning techniques can enhance predictive accuracy and robustness. The study covers regulatory requirements under frameworks like Basel III, implementation challenges, and performance measures that institutions use to evaluate AI systems. AI and machine learning technologies enhance data quality, automate workflows, strengthen compliance monitoring, and increase model precision, helping financial institutions streamline their CCAR processes while ensuring greater accuracy and transparency. However, banks still face significant hurdles in making AI models explainable, addressing bias issues, and managing systemic risks. The research shows that AI-driven approaches often perform better than conventional methods in accuracy and speed, but institutions need to balance innovation with regulatory compliance and risk management. This review provides insights for bank executives, risk managers, regulators, and researchers working to understand how AI is reshaping financial risk management and what it means for banking stability. Keywords: Artificial Intelligence; Stress Testing; Risk Assessment; Financial Institutions; Machine Learning; Regulatory Compliance 1. Introduction The global financial services industry stands at a critical inflection point, where traditional risk management approaches are proving increasingly inadequate in addressing the complexity and velocity of modern financial markets. The 2008 financial crisis exposed fundamental weaknesses in conventional risk modeling techniques, revealing how linear statistical models and historical pattern recognition failed catastrophically when faced with unprecedented market conditions and interconnected systemic risks [1]. This moment triggered a global regulatory response that has fundamentally reshaped expectations for financial risk assessment, with institutions now required to demonstrate resilience across multiple stress scenarios while maintaining real-time visibility into their risk exposures. Against this backdrop, Artificial Intelligence has emerged not merely as a technological enhancement but as an essential capability for financial institutions seeking to navigate increasingly complex regulatory requirements, manage growing data volumes, and maintain competitive positioning in rapidly evolving markets. AI technologies help solve these problems by allowing banks to process huge amounts of different types of data, spot complex patterns, and make more accurate risk predictions [2]. The combination of better computing power, advanced algorithms, and more available data has created a situation where AI-driven risk management isn't just helpful but increasingly necessary for banks to stay competitive and meet regulatory requirements. World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2510 Today's financial institutions work under intense regulatory oversight, changing market conditions, and growing connections between different parts of the financial system. The FDIC and OCC require covered institutions with assets over $250 billion to conduct stress tests using specific scenarios, with results providing forward-looking insights for supervisors [3]. In 2024, all 31 major banks including JPMorgan Chase and Goldman Sachs passed the Federal Reserve's annual stress test exercise [4]. Traditional econometric models, while mathematically sound, often can't capture the nonlinear relationships and dynamic connections that characterize modern financial systems. Using AI in financial risk management represents a shift from rule-based to data-driven approaches, letting institutions move beyond just recognizing historical patterns toward predicting and prescribing actions [5]. This change involves not just technology improvements but also fundamental changes in how organizations are structured, what talent they hire, and how they make strategic decisions. 2. Theoretical Framework and AI Technologies Overview AI applications in financial risk management combine computational intelligence with established financial theory and risk management principles. Modern portfolio theory, option pricing models, and capital asset pricing frameworks provide the mathematical foundation that AI algorithms use to build better predictive capabilities [6]. When traditional approaches merge with machine learning methods, they create hybrid models that use both theoretical knowledge and pattern recognition from data. Machine learning algorithms in finance generally fall into supervised learning, where models learn from labeled historical data to predict future outcomes, unsupervised learning that finds hidden patterns in data without predefined answers, and reinforcement learning that learns optimal actions through trial and error [7]. Each approach offers different strengths for various risk management tasks. Neural networks and deep learning represent the most sophisticated AI technologies currently used in financial risk assessment [8]. These systems can model complex, nonlinear relationships between multiple variables that traditional statistical methods struggle to capture. Deep learning architectures like recurrent neural networks excel at processing sequential data such as time series of market prices, while convolutional neural networks can analyze patterns in structured data representations [9]. The infrastructure needed to support AI-driven risk management requires substantial computational resources, data storage capabilities, and specialized software platforms. Financial institutions must invest in high-performance computing systems, establish robust data governance frameworks, and develop the technical expertise needed to implement and maintain these systems. The integration of AI technologies with existing risk management systems presents both technical and operational challenges that institutions must carefully manage. 3. AI Applications in Stress Testing 3.1. Scenario Generation and Modeling AI transforms how financial institutions create and analyze stress test scenarios by automating the generation of multiple economic conditions and market situations. Firms can update their stress-testing capability by harnessing automated scenario generation, allowing banks to move beyond the limited set of regulatory scenarios to explore a broader range of potential future conditions. Machine learning algorithms analyze historical economic data, market patterns, and macroeconomic indicators to generate thousands of plausible scenarios that capture both typical and extreme market conditions [10]. Traditional scenario generation relied heavily on expert judgment and historical precedent, which often missed emerging risks or novel combinations of economic factors. AI-driven approaches use generative models to create scenarios that include complex interactions between different economic variables, such as how interest rate changes might interact with unemployment rates, housing prices, and corporate default rates simultaneously [11]. These models can also incorporate real-time data feeds to ensure scenarios reflect current market conditions rather than being based solely on historical relationships. 3.2. Monte Carlo Simulations and Advanced Analytics Monte Carlo simulation techniques powered by AI enable financial institutions to run millions of possible outcomes for their portfolios under different stress scenarios [12]. AI enhances traditional Monte Carlo methods by using machine World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2511 learning to identify the most relevant variables to include in simulations and optimizing the sampling process to focus computational resources on the most probable and impactful scenarios. Advanced analytics platforms use AI to process the massive amounts of data generated by Monte Carlo simulations, identifying patterns and relationships that human analysts might miss [13]. These systems can automatically flag scenarios that produce unexpected results, helping risk managers understand which combinations of factors pose the greatest threats to their institutions. The speed and scale of AI-powered simulations allow banks to test their resilience against many more scenarios than traditional methods would allow [14]. 3.3. Model Validation and Back-testing AI technologies improve the validation and testing of stress test models by automatically comparing model predictions against actual outcomes across different time periods and market conditions [15]. Machine learning algorithms can identify when models are becoming less accurate and suggest adjustments or replacements before problems become serious. This continuous monitoring approach helps banks maintain model performance and meet regulatory expectations for model governance. Back-testing processes benefit from AI's ability to analyze large datasets and identify subtle patterns in model performance that traditional statistical tests might miss. AI systems can evaluate how well models perform across different market regimes, economic cycles, and portfolio compositions, providing more comprehensive assessments of model reliability than conventional approaches [16]. 3.4. Real-time Stress Testing Capabilities AI enables financial institutions to move from periodic stress testing to continuous, real-time risk monitoring. Machine learning models can process streaming market data, news feeds, and economic indicators to continuously update stress test results as conditions change [17]. This real-time capability allows banks to respond more quickly to emerging risks and adjust their strategies based on current rather than historical information. Real-time stress testing systems use AI to prioritize which scenarios to run based on current market conditions and emerging risks. When market volatility increases or new economic data becomes available, these systems automatically adjust their focus to the most relevant stress scenarios, ensuring that computational resources are used efficiently while maintaining comprehensive risk coverage. 4. Risk assessment through ai technologies 4.1. Credit Risk Modeling and Prediction Machine learning algorithms are suitable for dealing with various risk types banks face, with academic literature focusing on applying ML in credit risk management [18]. AI-powered credit risk models analyze borrower data, transaction patterns, and external economic factors to predict default probabilities more accurately than traditional scoring methods. These models can incorporate alternative data sources such as social media activity, spending patterns, and behavioral indicators that provide insights beyond traditional credit bureau information. Machine learning approaches excel at identifying complex relationships between different risk factors that linear models might miss [19]. For example, AI models can detect how combinations of employment history, geographic location, spending patterns, and economic conditions interact to influence credit risk in ways that traditional models would not capture. This enhanced predictive power helps banks make better lending decisions and price credit risk more accurately. 4.2. Market Risk Analytics and Volatility Forecasting AI technologies transform market risk management by providing more accurate predictions of price movements, volatility patterns, and correlation changes across different financial instruments [20]. Deep learning models can analyze high-frequency trading data, news sentiment, and macroeconomic indicators to forecast market movements with greater precision than traditional econometric models. Volatility forecasting benefits particularly from AI's ability to process multiple data sources simultaneously and identify regime changes in market behavior [21]. Neural networks can detect when market conditions are shifting from calm to World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2512 turbulent periods earlier than traditional volatility models, giving traders and risk managers more time to adjust their positions and hedging strategies. 4.3. Operational Risk Detection and Management Operational risk management uses AI to identify potential failures in processes, systems, and human behavior before they cause significant losses [22]. Machine learning algorithms analyze transaction patterns, system logs, and employee behavior to detect anomalies that might indicate fraud, system failures, or process breakdowns. These systems can process much larger volumes of data than human analysts and identify subtle patterns that might indicate emerging operational risks. Studies shows that AI-powered operational risk systems learn from historical incidents to improve their detection capabilities over time [23]. As new types of operational risks emerge, these systems can adapt their monitoring approaches to identify similar patterns in the future. This adaptive capability is particularly valuable in today's rapidly changing technological environment where new operational risks constantly emerge. 4.4. Liquidity Risk Assessment and Monitoring Machine learning models can detect liquidity risk by analyzing relationships between credit portfolio quality, assets, funding strategies, and market conditions, solving limitations in traditional Basel frameworks [24]. AI enhances liquidity risk management by analyzing complex relationships between funding sources, market conditions, and institutional cash flows. These models can predict when liquidity stress might occur and identify which funding sources might become unavailable during different market scenarios. Liquidity forecasting models use AI to analyze patterns in customer behavior, market liquidity conditions, and regulatory requirements to predict future cash flow needs [25]. These models help banks optimize their liquidity buffers and funding strategies by identifying when they might need additional liquidity and what sources would be most reliable during stress periods. 5. Regulatory Framework and Compliance 5.1. Basel III Requirements and AI Integration Basel III is an internationally agreed set of measures developed by the Basel Committee on Banking Supervision in response to the financial crisis of 2007-09, aiming to strengthen regulation, supervision and risk management of banks [26]. Financial institutions must ensure their AI-powered risk management systems comply with Basel III requirements for capital adequacy, liquidity coverage, and leverage ratios. AI models used for regulatory capital calculations must meet strict validation standards and provide transparent, auditable results that regulators can understand and verify. The integration of AI with Basel III frameworks requires careful attention to model governance, documentation, and validation processes [27]. Banks must demonstrate that their AI models produce consistent, reliable results that meet regulatory standards for accuracy and stability. This includes showing that models perform well across different market conditions and time periods, and that they don't introduce new sources of model risk. 5.2. Model Governance and Validation Standards Regulatory authorities require financial institutions to maintain rigorous governance frameworks for AI models used in risk management and regulatory reporting [28]. These frameworks must address model development, validation, implementation, and ongoing monitoring processes. AI models present unique challenges for traditional model validation approaches because they often operate as "black boxes" that are difficult for humans to interpret and understand. Model validation for AI systems requires new approaches that can assess model performance while accounting for the complexity and opacity of machine learning algorithms [29]. Validators must develop techniques for testing model stability, identifying potential biases, and ensuring that models behave appropriately under different conditions. This includes stress testing the models themselves to ensure they remain reliable during market disruptions. World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2513 5.3. Supervisory Expectations and Guidelines Banking supervisors worldwide are developing specific guidance for the use of AI in risk management and regulatory compliance [30]. These guidelines address expectations for model development, validation, governance, and risk management practices when using AI technologies. Supervisors emphasize the need for institutions to maintain human oversight of AI systems and ensure that they can explain and justify decisions made by these systems. Regulatory guidance increasingly focuses on ensuring that AI systems don't introduce new sources of systemic risk or create unfair outcomes for customers [31]. Banks must demonstrate that their AI systems operate fairly across different customer groups and don't perpetuate or amplify existing biases in lending or other financial services. This requires ongoing monitoring and testing of AI systems to identify and address potential bias issues. AI and ML technologies must ensure greater accuracy and transparency in financial processes, with regulators requiring financial institutions to explain how their AI models make decisions, particularly for models used in lending, regulatory capital calculations, and other critical functions [32]. This creates challenges because many advanced AI models, especially deep learning systems, operate in ways that are difficult for humans to understand and explain. Financial institutions are developing new approaches to make AI models more interpretable, including using simpler models where possible, creating explanation systems that can describe model decisions in human-understandable terms, and developing visualization tools that help analysts understand how models work [33]. These efforts must balance the need for model performance with regulatory requirements for transparency and explainability. 6. Performance Evaluation and Comparative Analysis 6.1. Accuracy Metrics and Benchmarking Evaluating AI performance in financial risk management requires sophisticated metrics that go beyond traditional statistical measures [34]. Banks use various accuracy measures including prediction error rates, area under the receiver operating characteristic curve, and precision-recall metrics to assess how well AI models perform compared to traditional approaches. These metrics must be evaluated across different time periods, market conditions, and portfolio segments to provide comprehensive performance assessments. Benchmarking AI models against traditional risk management approaches reveals significant performance improvements in many areas. AI models typically show better predictive accuracy for credit risk, more precise volatility forecasting for market risk, and improved detection rates for operational risk events [35]. However, performance advantages vary depending on the specific application, data quality, and market conditions being analyzed. 6.2. Computational Efficiency and Scalability AI systems often provide substantial improvements in computational efficiency compared to traditional risk management approaches. Machine learning models can process large datasets much faster than conventional statistical methods, enabling banks to run more comprehensive risk analyses in shorter time periods [36]. This efficiency gain becomes particularly important for real-time risk monitoring and high-frequency stress testing applications. Scalability represents another key advantage of AI systems, as they can handle growing data volumes and increasingly complex risk calculations without proportional increases in computational resources. Traditional risk management systems often require significant additional infrastructure as banks grow or add new products, while AI systems can often scale more efficiently by leveraging cloud computing resources and optimized algorithms [37]. 6.3. Model Robustness Across Market Conditions Testing AI model performance across different market conditions reveals both strengths and weaknesses compared to traditional approaches. AI models often perform better during normal market conditions but may face challenges during extreme market stress when historical patterns break down [38]. This requires careful evaluation of how well AI models generalize to conditions they haven't seen before and whether they maintain reasonable performance during market disruptions. Robustness testing involves exposing AI models to various market scenarios, including historical crises, hypothetical extreme events, and gradual changes in market structure [39]. Banks must evaluate whether their AI models maintain World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2514 acceptable performance across these different conditions and develop backup approaches for situations where AI models might not perform reliably. 6.4. Validation Framework Assessment AI/ML empowers financial institutions to streamline their CCAR processes while ensuring greater accuracy and transparency, reducing manual effort through comprehensive validation frameworks [40]. Comparing validation approaches for AI versus traditional models shows that AI systems require more sophisticated testing methods but can often provide more comprehensive validation results. AI models can be tested against larger datasets and more scenarios than traditional models, potentially providing better assurance of model reliability. Validation frameworks for AI models must address unique challenges such as model interpretability, bias detection, and performance stability over time [41]. These frameworks often require new statistical techniques and validation approaches that can handle the complexity of machine learning algorithms while meeting regulatory requirements for model governance and oversight. 7. Challenges and Limitations Financial institutions face significant obstacles when implementing AI for stress testing and risk assessment. Model interpretability remains a major concern, as regulators and internal stakeholders need to understand how AI systems make decisions [42]. Many advanced machine learning models operate as "black boxes," making it difficult to explain why they produce specific results or recommendations. This lack of transparency can create compliance issues and reduce confidence in model outputs during critical decision-making processes. Data quality and availability present ongoing challenges for AI implementation in risk management. AI models require large amounts of high-quality, relevant data to perform effectively, but financial institutions often struggle with incomplete datasets, inconsistent data formats, and data that may not be representative of future conditions. Poor data quality can lead to biased or inaccurate model results, potentially creating new risks rather than reducing them [43]. Algorithmic bias represents another serious concern, as AI models can perpetuate or amplify existing biases present in historical data [44]. This can lead to unfair treatment of certain customer groups or systematic errors in risk assessment that could have significant financial and reputational consequences. Banks must invest substantial resources in bias detection and mitigation strategies to ensure their AI systems operate fairly and comply with regulatory requirements. The integration of AI systems with legacy banking infrastructure creates technical and operational challenges. Many financial institutions operate on older technology platforms that weren't designed to support modern AI applications [45]. Upgrading these systems while maintaining operational stability and regulatory compliance requires careful planning, significant investment, and specialized technical expertise that may be difficult to acquire and retain. 8. Future Directions and Research Gaps The future of AI in financial risk management points toward more sophisticated, integrated approaches that combine multiple AI technologies to address complex risk challenges. Explainable AI represents a critical area of development, as financial institutions need AI systems that can provide clear, understandable explanations for their decisions while maintaining high performance levels. Research in this area focuses on developing new algorithms and techniques that balance model accuracy with interpretability requirements. Real-time risk monitoring and adaptive stress testing represent emerging applications that could transform how banks manage risk [46]. Future AI systems may continuously adjust their risk assessments based on streaming market data, news events, and other real-time information sources. This could enable banks to respond more quickly to emerging risks and adjust their strategies dynamically as conditions change. The integration of alternative data sources, including satellite imagery, social media sentiment, and economic indicator feeds, offers new opportunities for improving risk prediction accuracy [47]. AI systems can potentially process and analyze these diverse data sources to identify risk patterns that traditional approaches would miss. However, this also creates new challenges around data privacy, regulatory compliance, and model validation. Research gaps remain in understanding how AI models perform during unprecedented market conditions and how they might contribute to or help prevent systemic financial risks. The interconnected nature of modern financial systems World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2515 means that widespread adoption of similar AI models could potentially create new sources of systemic risk if these models fail or behave unexpectedly during stress periods. 9. Conclusion The implementation of Artificial Intelligence technologies in financial stress testing and risk assessment represents a paradigm shift that is fundamentally transforming the banking industry. The evidence presented throughout this comprehensive review demonstrates that AI-powered approaches consistently outperform traditional methods across multiple dimensions, including predictive accuracy, computational efficiency, and the ability to process complex, highdimensional datasets. Financial institutions that have successfully integrated AI into their risk management frameworks report significant improvements in their ability to identify emerging risks, generate more comprehensive stress test scenarios, and respond dynamically to changing market conditions. However, these technological advances come with new responsibilities and challenges that require careful consideration and strategic planning. The regulatory landscape surrounding AI in financial services continues to evolve rapidly, with supervisors worldwide developing new frameworks and expectations for AI governance, model validation, and risk management. The tension between innovation and compliance presents ongoing challenges for financial institutions, as they must balance the competitive advantages offered by AI technologies with the need to meet increasingly stringent regulatory requirements for transparency, explainability, and fairness. The successful navigation of this regulatory environment requires not only technical expertise but also a deep understanding of supervisory expectations and the ability to demonstrate that AI systems enhance rather than compromise financial stability and consumer protection. Looking toward the future, the continued evolution of AI technologies promises even greater capabilities for financial risk management, including real-time adaptive stress testing, enhanced predictive accuracy through alternative data integration, and more sophisticated approaches to systemic risk assessment. However, realizing these benefits will require ongoing collaboration between financial institutions, technology providers, and regulatory authorities to address current limitations and ensure that AI adoption serves the broader public interest. The institutions that succeed in this transformation will be those that can effectively combine technological innovation with strong risk management practices, regulatory compliance, and a commitment to fair and transparent financial services. Recommendations Financial institutions embarking on AI implementation for stress testing and risk assessment should adopt a phased approach that prioritizes regulatory compliance and risk management from the outset. Organizations should begin by establishing robust data governance frameworks and investing in high-quality data infrastructure before implementing AI models, as the success of any AI initiative depends fundamentally on data quality and availability. Institutions should also prioritize the development of internal AI expertise through targeted hiring, training programs, and partnerships with technology providers, ensuring that they have the necessary skills to implement, validate, and maintain AI systems effectively. Additionally, banks should establish clear model governance frameworks specifically designed for AI systems, including validation methodologies that can handle the complexity and opacity of machine learning algorithms while meeting regulatory requirements. Regulatory authorities should continue developing comprehensive guidance for AI use in financial services while fostering innovation through regulatory sandboxes and pilot programs that allow institutions to test new approaches under supervisory oversight. Regulators should invest in their own AI expertise and technological capabilities to effectively supervise AI-powered financial institutions and assess the risks and benefits of these technologies. International coordination among regulatory authorities is essential to ensure consistent standards and prevent regulatory arbitrage, particularly as financial institutions increasingly operate across multiple jurisdictions with varying AI governance requirements. Furthermore, regulators should engage proactively with the industry to understand emerging AI applications and their potential implications for financial stability, consumer protection, and fair lending practices. The broader financial services ecosystem, including technology providers, academic researchers, and industry associations, should collaborate to address common challenges such as model interpretability, bias detection and mitigation, and the development of industry-wide standards for AI governance and validation. Technology providers should prioritize the development of explainable AI solutions that can meet regulatory requirements for transparency while maintaining high performance levels. Academic researchers should focus on addressing critical knowledge gaps, particularly around the behavior of AI models during unprecedented market conditions and the potential for AI adoption to create new sources of systemic risk. Industry associations should facilitate knowledge sharing and best World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2516 practice development, helping smaller institutions benefit from AI advances while ensuring that the entire financial system moves forward in a coordinated and responsible manner. This collaborative approach will be essential for realizing the full potential of AI in financial risk management while maintaining the stability and integrity of the global financial system. Compliance with ethical standards Disclosure of conflict of interest No conflict of interest to be disclosed. References [1] Nafiu A, Balogun SO, Oko-Odion C, Odumuwagun OO. Risk management strategies: Navigating volatility in complex financial market environments. [2] Ashta A, Herrmann H. Artificial Intelligence and fintech: An overview of opportunities and risks for banking, investments, and microfinance. Strategic Change. 2021 May;30(3):211-22. [3] Moshirian F. The global financial crisis and the evolution of markets, institutions and regulation. Journal of banking and Finance. 2011 Mar 1;35(3):502-11. [4] Hanson SG, Ivashina V, Nicolae L, Stein JC, Sunderam A, Tarullo DK. The evolution of banking in the 21st century: Evidence and regulatory implications. Brookings Papers on Economic Activity. 2024;2004(1):343-89. [5] Javaid HA. AI-driven predictive analytics in finance: Transforming risk assessment and decision-making. Advances in Computer Sciences. 2024 Jun 11;7(1). [6] Giudici P. Fintech risk management: A research challenge for Artificial Intelligence in finance. Frontiers in Artificial Intelligence. 2018 Nov 27;1:1. [7] Patel AA. Hands-on unsupervised learning using Python: how to build applied machine learning solutions from unlabeled data. O'Reilly Media; 2019 Feb 21. [8] Mashrur A, Luo W, Zaidi NA, Robles-Kelly A. Machine learning for financial risk management: a survey. Ieee Access. 2020 Nov 5;8:203203-23. [9] Chen JF, Chen WL, Huang CP, Huang SH, Chen AP. Financial time-series data analysis using deep convolutional neural networks. In2016 7th International conference on cloud computing and big data (CCBD) 2016 Nov 16 (pp. 87-92). IEEE. [10] Khunger A. DEEP Learning for financial stress testing: A data-driven approach to risk management. International Journal of Innovation Studies. 2022 Mar 22. [11] Costa CJ, Aparicio JT, Aparicio M. Socio-Economic Consequences of Generative AI: A Review of Methodological Approaches. arXiv preprint arXiv:2411.09313. 2024 Nov 14. [12] Monte Carlo simulation techniques powered by AI enable financial institutions to run millions of possible outcomes for their portfolios under different stress scenarios [13] Kibria MG, Nguyen K, Villardi GP, Zhao O, Ishizu K, Kojima F. Big data analytics, machine learning, and Artificial Intelligence in next-generation wireless networks. IEEE access. 2018 May 17;6:32328-38. [14] Rane N, Choudhary S, Rane J. Artificial Intelligence for enhancing resilience. Journal of Applied Artificial Intelligence. 2024 Sep 9;5(2):1-33. [15] Owen AE. AI-Driven Stress Testing Framework for Credit Portfolios using MCMC Simulation. [16] Rahmani AM, Rezazadeh B, Haghparast M, Chang WC, Ting SG. Applications of Artificial Intelligence in the economy, including applications in stock trading, market analysis, and risk management. IEEE Access. 2023 Jul 31;11:80769-93. [17] Chatzis SP, Siakoulis V, Petropoulos A, Stavroulakis E, Vlachogiannakis N. Forecasting stock market crisis events using deep and statistical machine learning techniques. Expert systems with applications. 2018 Dec 1;112:35371. World Journal of Advanced Research and Reviews, 2025, 26(03), 2509-2518 2517 [18] Leo M, Sharma S, Maddulety K. Machine learning in banking risk management: A literature review. Risks. 2019 Mar 5;7(1):29. [19] Goldstein BA, Navar AM, Carter RE. Moving beyond regression techniques in cardiovascular risk prediction: applying machine learning to address analytic challenges. European heart journal. 2017 Jun 14;38(23):1805-14. [20] Nafiu A, Balogun SO, Oko-Odion C, Odumuwagun OO. Risk management strategies: Navigating volatility in complex financial market environments. [21] Yadava A. The impact of AI-driven algorithmic trading on market efficiency and volatility: Evidence from global financial markets. Information Sciences. 2024 Dec;36(3):102015. [22] Savchenko M. The impact of Artificial Intelligence on risk management in the operational activities of financial institutions. Development Management. 2024 Dec 17;4(23):45-54. [23] Sundaramurthy SK, Ravichandran N, Inaganti AC, Muppalaneni R. AI-powered operational resilience: Building secure, scalable, and intelligent enterprises. Artificial Intelligence and Machine Learning Review. 2022 Jan 8;3(1):1-0. [24] Leo M, Sharma S, Maddulety K. Machine learning in banking risk management: A literature review. Risks. 2019 Mar 5;7(1):29. [25] Nanda AS. How to Implement Predictive Analytics in the Cash Management Process of Small and Medium Banks. Journal ID. 2025;9471:1297. [26] Isebor JE. The future of international banking regulations in response to the financial crisis of 2007/2009: after basel III then what next?. Available at SSRN 2429934. 2014 Apr 27. [27] Paleti S. Data Engineering for AI-Powered Compliance: A New Paradigm in Banking Risk Management. European Advanced Journal for Science and Engineering (EAJSE)-p-ISSN 3050-9696 en e-ISSN 3050-970X. 2024 Dec 20;2(1). [28] De Almeida PG, dos Santos CD, Farias JS. Artificial Intelligence regulation: a framework for governance. Ethics and Information Technology. 2021 Sep;23(3):505-25. [29] Bücker M, Szepannek G, Gosiewska A, Biecek P. Transparency, auditability, and explainability of machine learning models in credit scoring. Journal of the Operational Research Society. 2022 Jan 2;73(1):70-90. [30] Aziz LA, Andriansyah Y. The role Artificial Intelligence in modern banking: an exploration of AI-driven approaches for enhanced fraud prevention, risk management, and regulatory compliance. Reviews of Contemporary Business Analytics. 2023 Aug;6(1):110-32. [31] Scherer MU. Regulating Artificial Intelligence systems: Risks, challenges, competencies, and strategies. Harv. JL and Tech.. 2015;29:353. [32] Rane N, Choudhary S, Rane J. Explainable Artificial Intelligence (XAI) approaches for transparency and accountability in financial decision-making. Available at SSRN 4640316. 2023 Nov 17. [33] Ahmad T, Katari P, Pamidi Venkata AK, Ravi C, Shaik M. Explainable AI: Interpreting Deep Learning Models for Decision Support. Advances in Deep Learning Techniques. 2024;4(1):80-108. [34] Khunger A. DEEP Learning for financial stress testing: A data-driven approach to risk management. International Journal of Innovation Studies. 2022 Mar 22. [35] Yazdi M, Zarei E, Adumene S, Beheshti A. Navigating the power of Artificial Intelligence in risk management: a comparative analysis. Safety. 2024 Apr 26;10(2):42. [36] ¬¬¬Shen Q. AI-driven financial risk management systems: Enhancing predictive capabilities and operational efficiency. Applied and Computational Engineering. 2024 Jul 25;69:134-9. [37] Hammad A, Abu-Zaid R. Applications of AI in Decentralized Computing Systems: Harnessing Artificial Intelligence for Enhanced Scalability, Efficiency, and Autonomous Decision-Making in Distributed Architectures. Applied Research in Artificial Intelligence and Cloud Computing. 2024;7:161-87. [38] Ahmed AA, Abdullahi AU, Gital AY, Dutse AY. Application of Artificial Intelligence in Supply Chain Management: A Review on Strengths and Weaknesses of Predictive Modeling Techniques. Scientific Journal of Engineering, and Technology. 2024 Oct 11;1(2):1-8.