The banking sector in the Middle East is undergoing rapid transformation, driven by digital acceleration, rising supervisory expectations and a growing dependence on complex decision models. These models have quietly become the core engines of decision-making — influencing everything from credit underwriting and capital planning to fraud detection, customer engagement. The influence is now expanding further, adding new dimensions to model risk with the rapid rise in use of machine learning (ML) and AI techniques, especially in data-rich and customer-facing use cases.
As this reliance and use deepens, so does the need for robust model risk management (MRM). No longer just a regulatory requirement, MRM is emerging as a strategic priority for banks seeking to manage complexity, mitigate risk and remain resilient in a fast-changing environment as well as maintain competitive advantage.
We engaged with 23 leading banks across the United Arab Emirates (UAE), Saudi Arabia, Oman and Qatar to conduct a first-of-its-kind survey for better understanding how the MRM landscape is evolving — and the challenges institutions continue to face. This article is more than just a summary of findings, it is a mirror and a map reflecting the shared intent behind that effort — to benchmark MRM practices, spark dialogue, and contribute to the region’s journey towards more resilient model governance. The report is designed to help banks and risk leaders reflect on their current maturity, recognize shifting regulatory expectations, and take meaningful steps to shape the future of MRM in the region.
Key findings
-
Model governance
Banks are making clear progress in formalizing model governance framework including establishment of model oversight committee.
-
Model risk tiering
61% of the surveyed banks have adopted structured model risk tiering frameworks with model materiality, usage, and complexity identified as preferred drivers.
-
MRM inventory and automation
83% of surveyed banks maintain a formal model inventory with coverage still expanding.
-
Policy maturity
Despite existing standards and frameworks, the absence of strong governance around data and infrastructure continues to hinder policy maturity across the model lifecycle.
MRM resourcing
Banks vary in their MRM resourcing models—some scale internally, while others rely heavily on outsourcing or operate with limited internal capacity.
-
Model validation
Banks are strengthening and restructuring validation frameworks to address constraints around model complexity, data availability, and internal resourcing.
-
Risk appetite and capital assessment
Growing adoption of model risk appetite and capital assessment frameworks reflect increasing maturity and integration of model risk in broader risk management.
-
Regulatory readiness
Many banks are progressing to ensure compliance with supervisory expectations over next twelve months in the regions with formal MRM regulations, while others are proactively aligning with leading industry practices.
