An Unparalleled Framework for AI and Model Risk Management

Major financial institutions use models to manage risks, create investment strategies, increase sales, and improve process efficiency. Better models produce better results. Sound modeling practice can help to improve institutional compliance with regulatory model risk requirements (OCC 2011-12/FRB SR 11-7, FHFA AB 2013-07 and AB 2022-02, FIL-22-2017).

With this in mind, we offer you a holistic model risk management framework based on three key elements: governance, model life cycle, and model validation (more details in lists below). The same principles can be customized and applied to manage AI and Machine Learning model risk.

In addition, our sophisticated approach can reveal hidden developmental errors in existing models, or how they may be improperly used.

Model Risk Management Framework

Governance

Board, Management and MRM

Policy and procedures
Model identification, classification, and inventory
Performance monitory
Issue tracking
Reporting
Training

Risk Mitigation in Model Life Cycle

Model Owners

Development/Acquisition
Implementation
Testing
Performance tracking
Change management
Documentation

Independent Model Validation

MRM Group

Data input
Assumptions
Theory, math, and code
Outputs
Risk and limitation

Proven Success Story: Increase Benchmark Capacity from 5% to 99%

We bring hands-on experience in translating regulatory guidance to the policy and implementing internal guidance. Our services are based on proven success in creating, designing, implementing, and maintaining a model risk management framework for both the second (risk) and the third lines (audit).

In one case, we designed and implemented the MRM framework from the ground up and built in-house capacity that enabled the company to increase benchmarking coverage from less than 5% to overall 99% on balance sheets and to expand frequency from every three years to annually, quarterly, or as needed. The framework substantially improved business effectiveness, efficiency, transparency (understanding black boxes), and flexibility while reducing risk and cost.