Fannie Mae and Freddie Mac Set New AI Standards for Mortgage Lenders
Fannie Mae and Freddie Mac have introduced new AI governance frameworks for mortgage lenders, requiring compliance with strict rules on AI and machine learning use in loan origination and servicing by August 2026. Freddie Mac’s guidelines, already effective, mandate detailed operational controls and audits, while Fannie Mae’s principles-based framework applies broadly across AI systems, emphasizing fairness and accountability under fair lending laws.
Fannie Mae and Freddie Mac are enforcing new AI governance requirements for mortgage lenders, marking a shift from experimental use to mandatory compliance in loan processing. The rules apply to all AI or machine learning systems used in origination or servicing, including vendor-provided tools, and directly impact lenders’ ability to sell loans to these institutions. Freddie Mac implemented its guidelines first, issuing Guide Bulletin 2025-16 on December 3, 2025, which took effect March 3, 2026. The bulletin imposes detailed governance, audit, and security requirements, codified in Section 1302.8 of its Single Family Seller/Servicer Guide. It mandates operational controls, regular audits, and explicit indemnification obligations for AI systems, ensuring accountability across the loan lifecycle. Fannie Mae followed with Lender Letter LL 2026-04 in April 2026, establishing a principles-based governance framework for AI and machine learning in mortgage lending. The framework applies to all approved sellers and servicers, regardless of whether the AI tools are internally developed or vendor-provided. Compliance is required by August 6, 2026, and the rules extend beyond underwriting to include pricing, fraud detection, and loss mitigation. Both frameworks align in emphasizing fairness and compliance with laws like the Equal Credit Opportunity Act and Regulation B, though Freddie Mac’s approach is more prescriptive. Lenders selling to both institutions must meet the stricter requirements of each. The changes reflect broader regulatory concerns about AI’s impact on transparency and accountability in mortgage lending, where traditional rule-based models are being replaced by complex, often opaque algorithms. The shift underscores how AI is now integral to mortgage operations, from borrower intake to servicing, but also introduces new compliance challenges. Lenders must now document AI decision-making processes, ensure auditability, and mitigate risks of bias or discrimination. Failure to comply could restrict access to the secondary mortgage market, where Fannie Mae and Freddie Mac play a central role.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.