Federal: AFC Letter in Response to Request for Information on AI Risks and Modernization in Financial Services

The Honorable Maxine Waters
Ranking Member
United States House of Representatives Committee on Financial Services
2221 Rayburn House Office Building
Washington, D.C. 20515

Re:       Letter in Response to Request for Information on AI Risks and Modernization in Financial Services

Dear Ranking Member Waters,

On behalf of the American Fintech Council (AFC) and its members,  I write in response to the Request for Information (RFI) regarding the use of artificial intelligence (AI) in the financial services and housing industries. AFC appreciates the Committee’s sustained bipartisan leadership on this issue, from the creation of the Fintech and AI Task Forces in 2019 through the Bipartisan Working Group on AI’s 2024 staff report, and we welcome the opportunity to contribute our members’ perspective as Congress considers whether and how to modernize the federal framework governing AI in financial services.

A standards-based organization, AFC is the largest and most diverse trade association representing financial technology companies and innovative banks. On behalf of more than 150 member companies and partners, AFC promotes a transparent, inclusive, and customer-centric financial system by supporting responsible innovation in financial services and encouraging sound public policy. AFC members foster competition in consumer finance and pioneer products that better serve underserved consumer segments and geographies. They also share a commitment to the responsible development and deployment of emerging technologies, including AI, in a manner that promotes consumer protection, operational resilience, and regulatory accountability.

AFC has been significantly engaged in the AI issue over the past two years. In August 2024, AFC submitted a comment letter to the Department of the Treasury on the uses, opportunities, and risks of AI in financial services, articulating the risk-based, context-specific approach to AI regulation that continues to guide our advocacy.  AFC has since supported a series of bipartisan measures before this Committee, including the Unleashing AI Innovation in Financial Services Act in August 2025, a slate of AI-related bills considered at the Committee's December 2025 markup, and further legislation considered at the Committee's May 2026 markup, including the Bank Fraud Technology Advancement Act, the GUARD Act, and the FUTURES Act.  AFC was also a signatory to a May 2026 joint trade association letter welcoming structured public-private collaboration on frontier AI model security.  

This letter draws on that record to respond to the RFI's questions below, organized by topic area, and we confirm that AFC has not submitted supplemental data or reports separate from this letter.

I. AFC Recommends a Risk-Based, Context-Specific Framework for AI Regulation

Fundamentally, AFC recognizes that AI is a tool that, when responsibly deployed, expands and improves existing financial products and services rather than replacing the judgment, accountability, and risk management already required of regulated institutions. Our advocacy on this topic has centered on the importance of developing a unified, risk-based federal framework for AI in financial services. AFC's central recommendation, since our original 2024 comment letter to the Treasury Department, has been that AI regulation in financial services should be risk-based and context-specific: when AI enables an existing financial product or service to reach more consumers, the existing regulatory framework for similarly situated products is generally sufficient; when AI is used to create a wholly new product or service, a distinct regulatory framework is warranted.  AFC continues to believe this framework is the right lens for the Committee's ongoing work to review model risk management, explainability, and deployment controls.

As such, AFC has recommended that financial institutions integrate AI into their existing third-party and model governance structures including regular audits, model validation, and clear accountability mechanisms rather than treating AI as requiring an entirely separate compliance regime, and has pointed to the National Institute of Standards and Technology’s (NIST) AI Risk Management Framework as a useful reference point for that work.  

Further, as Congress and regulators engage on the AI issue, it is crucial that they recognize how existing laws and regulations cover AI use cases. Much of the financial regulatory framework was established in a technology-neutral manner, meaning that it applies to activities regardless of the technology used or which entity pursues them. In the case of AI, this technology-neutral approach is crucial. At the state-level, AFC has seen legislatures move quickly to adopt legislation that merely creates duplicative or conflicting requirements with established laws. While AFC respects the intent of these state legislatures to pursue legislation that they believe will help citizens, it is critical that they take a pragmatic approach to policymaking and ensure that their laws will not cause further legal uncertainty. To that end, AFC was encouraged by the House resolution at the Committee's December 2025 markup affirming that existing consumer protection and anti-discrimination laws continue to apply to AI-driven decisions, and calls for any AI framework to be tailored so that it does not disproportionately burden smaller institutions.  AFC believes this same principle should guide the Committee's approach to Question 10: existing fair lending and adverse action requirements already establish the relevant legal baseline, and the priority should be ensuring institutions of all sizes can meet it.

II. AFC Recommends a Unified Federal Framework for AI Regulation

As previously noted, AFC has long advocated for a unified approach to AI regulation. Simply put, AI tools inherently lack geographic constraints, and therefore, should be regulated at the federal level. In our 2024 letter, AFC raised concerns that a state-by-state patchwork of AI regulation, citing Colorado's Concerning Consumer Protections in Interactions with Artificial Intelligence Systems as an example, creates compliance challenges for financial institutions operating across multiple jurisdictions and can create unnecessary barriers to entry for smaller institutions.  AFC recommended that Treasury and its fellow financial regulators work toward a more unified federal approach that provides clear “rules of the road” while preserving the risk-based, context-specific principles described above.  Further, a state-by-state approach to regulating AI naturally increases the risk of consumers misunderstanding, and responsible innovators misapplying, various state requirements. Ultimately, this could lead to consumers misidentifying the protections that they are entitled to in a given state, undercutting their trust in the government and financial institutions serving them. Given the Congress’ unique role, we would encourage the Committee to weigh the importance of establishing a unified federal regulatory framework for AI, especially as it considers Question 13's inquiry into potential amendments to federal data privacy law given the interaction between state AI and privacy regimes.

Relatedly, AFC recognizes the importance of ensuring that state and federal policymakers engage with their international counterparts to understand how AI regulatory frameworks outside of the U.S. interact with the U.S. regulatory framework for AI. Again, given that AI tools inherently lack geographic limitations, and U.S. financial services companies leveraging AI tools remain significantly engaged across the globe, it is crucial that regulatory frameworks regarding AI use cases in financial services are sufficiently aligned to avoid conflicting or incongruent requirements that would cause confusion for consumers and industry participants, as well as undercut domestic regulatory frameworks by enabling regulatory arbitrage and extraterritorial regulation.

III. AFC Recommends Regulatory Clarity on Less Discriminatory Alternatives in Fair Lending

In our August 2024 letter, AFC highlighted Less Discriminatory Alternatives (LDAs) as an area where the CFPB's Fair Lending Report identified a meaningful opportunity to modernize fair lending practices through the use of AI technologies.  AFC echoed a joint statement from Consumer Reports and the Consumer Federation of America calling for regulatory clarity regarding the expectation that financial institutions search for and implement less discriminatory algorithms in credit underwriting and pricing.  AFC continues to believe that encouraging the development and appropriate use of LDAs, where available, can help create a more equitable financial landscape while preserving industry efficiency and innovation, and we would encourage the Committee to consider how any explainability framework it advances can support that goal.

IV. AFC Recommends Expanding Smaller Institutions' Access to AI-Powered Compliance and Fraud Tools

AFC has consistently supported measures that expand smaller institutions' access to AI-powered compliance and fraud-fighting tools. In our letter supporting H.R. 8671, the Bank Fraud Technology Advancement Act of 2026, at the Committee's May 2026 markup, we noted that smaller banks and credit unions face the same fraud threat landscape as large institutions but often lack the resources to develop or procure cutting-edge detection and prevention capabilities.  We supported that bill's voluntary pilot program to expand community financial institutions' access to fraud-fighting technology as a collaborative, market-driven approach, and encouraged the Committee to ensure any such program is designed in consultation with fintech companies and technology providers, who are well-positioned to offer scalable, cost-effective solutions to institutions that lack the capacity to build these capabilities in-house.

This concern, that AI frameworks not disproportionately burden smaller institutions, is also reflected in the House resolution AFC supported in December 2025, discussed above, and we would encourage the Committee to apply that same principle to Question 21's inquiry regarding CDFIs' and MDIs' access to third-party AI tools relative to larger, better-resourced institutions.

V. AFC Recommends a Coordinated, Whole-of-Government Approach to Combatting AI-Enabled Fraud

AFC's record reflects sustained engagement on the use of AI to combat, and its potential misuse to perpetrate, financial crime. At the Committee's December 2025 markup, AFC supported H.R. 1734, the Preventing Deep Fake Scams Act, which would establish a task force to consult industry stakeholders, standardize core AI definitions, and recommend best practices and policy responses to AI-driven fraud and identity theft.  AFC also supported H.R. 2152, the AI PLAN Act, which directs the Departments of Treasury, Homeland Security, and Commerce to jointly produce a comprehensive assessment of the national and economic security risks posed by AI in financial crimes, along with legislative recommendations.  AFC emphasized that this whole-of-government approach, rather than legislating in advance of a full understanding of the threat landscape, is the right posture, and encouraged the Committee to ensure fintech companies and industry stakeholders are meaningfully consulted as the relevant agencies develop their findings.

At the Committee's May 2026 markup, AFC also supported H.R. 2978, the GUARD Act, addressing elder financial fraud and “pig butchering” scams.  AFC supported the bill's authorization for state and local law enforcement to use existing federal grant funding for training and enforcement related to financial fraud and elder fraud, noting that local officers are often the first responders to a fraud event and currently lack adequate resources. AFC also supported the bill's mandate for Treasury to study the state of fraud and scams and report on enforcement recommendations, citing FTC data showing that bank transfer and payment fraud alone accounted for $2.09 billion in consumer losses in 2024.

Separately, AFC's 2024 letter noted that AI systems, equipped to analyze large volumes of data, can help financial institutions monitor and flag suspicious activity in real time, strengthening Bank Secrecy Act compliance.  AFC believes this body of work, spanning AI-driven fraud task forces, an interagency financial-crime risk assessment, dedicated law enforcement resourcing, and AI's use in BSA/AML compliance, is directly responsive to the RFI's questions regarding how Congress and regulators can address the threat of AI-enabled financial crime, and we would encourage the Committee to continue advancing this slate of measures together.

VI. AFC Recommends Building on Structured Public-Private Collaboration on Frontier Model Security

AFC was a signatory, alongside other industry associations, to a May 2026 joint letter welcoming frontier AI developers' efforts to raise awareness and promote collaboration through initiatives such as Anthropic's Project Glasswing.  That letter specifically highlighted the Treasury Department's ongoing public-private AI-cybersecurity workstream — which has produced work on a financial services AI risk management framework, AI-enabled fraud, identity and authentication, explainability, “nutrition labels,” and a common AI lexicon — as a strong model of public-private collaboration for developing practical solutions and policy recommendations.

AFC continues to believe this kind of structured, voluntary collaboration among frontier AI developers, policymakers, and industry stakeholders, rather than a prescriptive mandate, is the right model for the questions the RFI raises regarding responsible AI frameworks and participation in initiatives like Project Glasswing, and we would encourage the Committee to build on Treasury's existing workstream, including the nutrition-label concept referenced in Question 17, rather than developing a new framework in parallel.

VII. AFC Recommends Advancing Regulatory Sandboxes and Supervisory Modernization Together

AFC has long supported H.R. 4801, the Unleashing AI Innovation in Financial Services Act, which would establish supervised AI Innovation Labs within federal financial regulatory agencies, providing regulated entities a safe environment to test and develop AI technologies in partnership with regulators.  AFC also supported the AI INSIGHT Act discussion draft at the Committee's December 2025 markup, which would direct Treasury to study best practices for data sharing in regulatory sandboxes focused on AI tools in financial services.  AFC believes this combination of supervised testing environments paired with a formal study of sandbox data-sharing practices is responsive to the RFI's questions regarding appropriate pre- and post-deployment controls, and we would encourage the Committee to advance both measures together.

Separately, AFC supported the FUTURES Act at the Committee's December 2025 and May 2026 markups, which would require financial regulators to assess whether their current technology impairs real-time supervisory oversight and to identify opportunities to modernize.  AFC has encouraged the Committee to ensure that review is conducted with meaningful industry input and to consider whether the FUTURES Act's mandate should explicitly encompass the CFPB, given its central supervisory role over many AFC members. AFC believes modernized supervisory technology is a necessary complement to the third-party oversight measures discussed in Questions 19 and 20 above.

* * *

AFC appreciates the Committee's continued engagement on these issues and the opportunity to contribute our members' perspective. We stand ready to serve as a resource to the Committee and its staff as this work continues, including through direct engagement with our members.

Sincerely,

Ian P. Moloney
Chief Policy Officer
American Fintech Council

CC:

The Honorable French Hill, Chairman, House Committee on Financial Services

‍