We are far beyond artificial intelligence being a future discussion for the mortgage industry. It is already embedded throughout the mortgage manufacturing process.
Mortgage companies of all sizes and structures utilize AI-enabled tools, both directly and indirectly, for document review, borrower communication, marketing, fraud detection, quality control, pricing support, servicing workflows, and internal productivity. In many cases, lenders are not intentionally “buying AI.” It exists and is deployed through the platforms and systems they already use every day.
For the largest financial institutions with extensive compliance, legal, model risk management, and technology teams, building internal AI governance structures is achievable on their own. For most independent mortgage bankers, community lenders, and mortgage brokers, that is a very different challenge.
That gap is exactly why the Mortgage Bankers Association’s Residential Board of Governors (RESBOG) identified the development and maintenance of an industry AI governance framework as one of its top priorities for 2026.
In partnership with RESBOG, a dedicated group of participants within MISMO’s AI Community of Practice has spent the past several months developing MISMO FRAME (Framework for Responsible AI in Mortgage Ecosystem). The initiative is designed specifically to help all mortgage companies build practical governance structures around the responsible use of AI.
Importantly, FRAME is not intended to create new regulations or compliance obligations. Existing laws and regulations already apply to lenders, irrespective of their deployment of AI. The challenge for many lenders is understanding how to operationalize those obligations when AI is involved.
FRAME is designed to provide practical guidance and implementation tools that mortgage companies can actually use. It incorporates broader standards such as the NIST AI Risk Management Framework but narrows the focus specifically to mortgage industry operations, risks, workflows, and regulatory expectations.
The framework will include practical implementation artifacts such as:
- AI system inventories
- Governance policy templates
- Vendor risk assessment structures
- AI system risk assessments
- Change management and monitoring guidance
- Generative AI governance considerations
One of the most important aspects of this effort is that it recognizes the operational reality of today’s mortgage ecosystem. Most mortgage companies are not building large language models internally. Instead, they are increasingly relying on AI-enabled capabilities embedded inside loan origination systems, CRM platforms, servicing technologies, QC systems, marketing tools, and other third-party ‘point’ solutions.
That means vendor oversight is becoming one of the most important components of AI governance. Even if a lender does not own the model, the lender still owns the outcome.
As the mortgage industry’s nonprofit standards organization, MISMO is uniquely positioned to lead this effort. MISMO operates “For the Industry, By the Industry,” bringing together lenders, brokers, servicers, vendors, GSEs, regulators, investors, and technology providers in a collaborative environment designed to create practical standards and frameworks that benefit the broader ecosystem and promote interoperability.
FRAME will be discussed in detail during MISMO's Spring Summit in Louisville, Ky., June 1–4, with broader availability to MISMO member companies following the event. Like many successful MISMO initiatives, the framework will evolve over time based on real-world implementation, industry feedback, and ongoing collaboration.
MISMO is also developing an AI Certification program that will help mortgage companies better assess vendor AI governance and oversight practices across the ecosystem.
The mortgage industry does not need every company to become an AI laboratory. It does, however, need every company to become a responsible operator of AI-enabled systems.
Anyone interested in learning more about the MISMO AI Framework initiative can contact me directly at [email protected].
