Freddie Mac is set to implement new guidelines requiring mortgage companies to establish a comprehensive artificial intelligence framework governing the internal use of AI technology. Effective March 3, 2026, these rules will mandate compliance from businesses utilizing AI and machine learning tools in relation to loans sold to or serviced on behalf of the government-sponsored enterprise.
The need for such guidelines has intensified amid the rapid growth of AI platforms from companies like OpenAI and other emerging competitors. Leaders in the mortgage industry have increasingly emphasized the necessity for lenders to create clear internal policies as they adopt new technologies, anticipating that this will lead to broader regulation across the sector. Freddie Mac’s update arrives at a crucial time, as the potential for both unintentional and deliberate misuse of AI technologies looms large.
As open-source platforms proliferate, many organizations are eagerly integrating AI into their daily operations. “Everybody’s starting to consume that and use it in their day-to-day,” said Joe Sorbello, product owner at mortgage software firm Lender Toolkit, which oversees underwriting platforms. However, the use of public AI tools raises concerns regarding data quality when employed for business purposes. “AI can do analysis, AI can feed back results,” Sorbello noted, “but you run the risk of AI learning something wrong.”
In December, Freddie Mac added a dedicated section for AI governance to its guidelines. Starting in March 2026, sellers and servicers will be required to demonstrate that they have processes in place for “mapping, measuring and managing AI risks” that are clear and effectively implemented. The guidelines also call for the introduction of “trustworthy” AI features, necessitating that mortgage companies develop processes to meet appropriate risk management standards.
With a strong focus on safety and compliance, Freddie Mac will expect mortgage sellers and servicers to monitor their AI tools for threats to data integrity and conduct audits to identify weaknesses that could hinder policy adherence. Additionally, the updated guidance emphasizes the importance of separating responsibilities within organizations to prevent conflicts of interest and ensure accountability. Documentation and communication of internal rules and roles concerning AI risk will be mandatory for all employees.
These guidelines make it clear that the ultimate responsibility for risk oversight—and any potential penalties—lies with the originators, even as many have historically relied on their technology partners to shoulder this burden. The update also highlights the necessity for thorough vetting of AI partners by lenders and servicers, particularly in defining the boundaries between automation and human oversight. “A lot of the lenders really rely on us, as the vendors, having those security protocols in place as it relates to security of data, use of data,” Sorbello added.
Freddie Mac’s move comes as mortgage businesses are increasingly embracing technological advancements while struggling to effectively integrate AI into their workflows. Research conducted by Arizent, the parent company of National Mortgage News, found that in 2025, 38% of companies were opting for a slow and cautious introduction of AI, with 2% imposing stricter limits.
Concerns over noncompliance in the absence of clear guidelines have made some industry leaders hesitant to adopt AI more aggressively. The introduction of explicit regulations, such as those from Freddie Mac, is expected to alleviate some of this apprehension and encourage wider adoption. “It’s going to cause people to ask the questions, which is a good thing,” Sorbello remarked. He emphasized that from a lender’s risk perspective, this clarity will help ensure that they continue to ask the right questions of their vendors, ultimately fostering a more responsible use of AI in the mortgage industry.
As Freddie Mac takes these crucial steps toward regulating AI use, it signals a shift in the mortgage industry towards more structured and responsible integration of emerging technologies, paving the way for potential growth and innovation in a sector increasingly influenced by advancements in artificial intelligence.
See also
Destination AI Expands 2026 Hospitality Summit to 1.5 Days Amid Surging Industry Demand
Nvidia Granted Approval to Sell H200 AI Chips to China, Targeting Billions in Revenue
Microsoft Unveils $17.5 Billion AI Investment in India to Expand Cloud Infrastructure and Workforce Training
Hyundai Reveals MobED Robot with AI for Stable Movement on Varied Surfaces
Chinese State Actors Leverage AI in Espionage: Anthropic’s Claude Code Discovered in Campaign
