Copyright forbes
Artificial Intelligence and machine learning possibilies abound in U.S. lending programs. A new study suggests artificial intelligence is poised to reshape how the U.S. government lends and manages its sprawling $5 trillion loan portfolio. While widespread adoption and full integration of AI tools across credit programs remain nascent, agencies have begun to lay the groundwork to enable more efficient and taxpayer-conscious lending processes. The report, from Ugur Koyluoglu of the management consulting firm Oliver Wyman, considers the potential for AI integration into U.S. credit programs — from underwriting and borrower services to risk management and compliance. The release follows the Trump administration’s July release of a federal “AI Action Plan.” A working group of current and former government officials sponsored the study to address the rise of programs that facilitate borrowing for homeownership, higher education, small businesses, and other public purposes. With outstanding balances totaling more than $5 trillion, these programs have grown to the point where the United States has become the nation’s largest lender — a scale requiring continual innovation and modernization to ensure borrower needs are balanced with taxpayer protection. A Pivotal Step For Federal Lending With agencies now facing uncertain budgets and reduced staffing levels, they must keep up with innovations being piloted in the private sector. The study sees AI not just as a tool to speed up work but also as a way to protect taxpayer money while improving results. Agencies have begun to use AI tools to predict risks and employ chatbots to help borrowers get answers. But a fuller embrace of AI could fundamentally change how staff do their jobs — helping them to be more efficient, lower loan losses, catch fraud, and provide better borrower service. MORE FOR YOU The study highlights five ways AI could be introduced or expanded: Deploy AI assistants or co-pilots to support credit analysis. Use machine learning to improve risk assessments and loan portfolio performance. Find and stop fraud more effectively. Make loan cost calculations more accurate and transparent. Increase the use of chatbots to help loan applicants. Using AI tools that allow humans and machines to collaborate is the most effective path forward. With AI learning from human feedback, such an approach heightens not only productivity but also the prospect that AI decisions adhere to legal and ethical standards. A Vision Rooted In Innovation, Tempered By History Despite operating in a very different budgetary and staffing environment than in recent years, agencies still must manage a portfolio that has nearly doubled since the 2008 financial crisis. The volume of new loans and loan guaranties continues at a robust pace given bipartisan support for loan programs and their typically low budgetary costs. Due to a variety of budget rules, nearly three-fourths of loan guaranty programs show no upfront budgetary cost. It is not hard to see how credit programs have arrived as a preferred form of assistance: They travel along a path of least budgetary resistance. Managing the government’s vast loan portfolio has proven to be difficult at times with surges in defaults, operational inefficiencies, and volatile outcomes plaguing some programs. The new AI paper acknowledges this legacy and considers AI as a potential solution to reducing such problems. For example, predictive AI can use borrower data and macroeconomic indicators to refine default risk models, while generative AI can automate loan documentation and reporting. The Department of Energy’s AI Toolbox, spotlighted as a model initiative, uses AI-driven content generation and cognitive engines to accelerate portfolio analysis and processing, demonstrating tangible gains in turnaround times and oversight effectiveness. Extending The Lending Frontier Federal credit programs can serve as a powerful proving ground for the White House's AI strategy by showcasing how AI can be used not only for operational efficiency but as a mechanism to shape national policy and unlock new public service capabilities. AI-guided lending could boost financing for sectors crucial to U.S. competitiveness such as infrastructure modernization and supply chain resilience. This approach also aligns with recently updated guidance from the Office of Management and Budget instructing agencies to design loan programs that complement, rather than compete against, private sector lending. Reflecting shifting national priorities, federal lending agencies may have a greater willingness to invest directly in high-impact economic sectors where private capital is hesitant. The study devotes substantial focus to AI governance: Making sure AI is used responsibly by creating safeguards so that AI systems do not make biased, harmful, or inaccurate decisions. In short, AI governance helps build public trust by keeping human judgment, fairness, and safety at the center of technological progress. Managing A Lender of All Resorts As the government’s loan portfolio continues to grow, questions persist in Congress and at the White House about the challenges faced such as fluctuating student loan cost projections and the unresolved conservatorship of Fannie Mae and Freddie Mac. As evidenced through those credit assistance efforts, well-intentioned lending can have unintended and long-lasting consequences. The AI report positions technology as a tool to help design and deliver programs to achieve policy goals while maintaining rigorous fiscal discipline. Effective AI deployment enables continuous portfolio monitoring and dynamic stress testing, empowering policymakers to adjust lending conditions and manage risks in near real-time — a capability increasingly adopted by the private sector. But such benefits depend on workforce readiness. Credit analysts and loan officers require new skills to interpret AI-generated insights effectively and to maintain appropriate levels of human interaction and judgment. The study provides a set of use cases translated into strategic guidance to accelerate the AI journeys of credit agencies. Key recommendations include: 1) Make AI adoption a priority and get going, 2) Develop an AI strategy and identify prioritized initiatives, 3) Set up key enablers such as governance, testing, controls, and talent, 4) Implement a robust roadmap, and 5) Continue to iterate and monitor AI use. Balancing Innovation And Prudence The ascent of AI offers a promising path forward, but it magnifies the need for transparent governance, ethical oversight, and interagency collaboration. The study’s vision combines optimism with realism: AI can make government lending smarter, faster, and fairer — but only if introduced thoughtfully and with sturdy guardrails. With the level of precision and accuracy expected in private sector loan systems, this initiative could transform the federal loan ecosystem from a fragmented patchwork into a resilient, data-driven system. If executed well, the nation’s largest lender stands to benefit from AI not just in operational efficiency, but in results — achieving precision, accountability, and continual improvement. Editorial StandardsReprints & Permissions
