Executive Summary
The Bank Policy Institute (“BPI”), and the law firm of Covington & Burling LLP (“Covington”), have developed this discussion draft to solicit input and views from relevant stakeholders on the appropriate regulatory framework for the use of artificial intelligence and machine learning (collectively, “AI”) in credit underwriting. BPI believes AI has great potential and can help banks improve access to consumer credit for underserved consumers and this discussion draft seeks to begin a dialogue among relevant stakeholders to help identify a path that regulators and industry can take to further these objectives. As a part of this effort, and based on the input provided by relevant stakeholders, BPI intends to release a final white paper with recommendations for modernizing regulatory approaches to the use of AI in credit underwriting.
AI offers a leap forward for the accuracy and fairness of decisions on consumer credit. AI can integrate and analyze richer data sets than conventional credit underwriting, and more accurately assess a consumer’s creditworthiness using factors (and combinations of factors) ordinarily not considered by conventional underwriting systems. This increased accuracy will benefit borrowers who currently face obstacles obtaining low-cost bank credit under conventional underwriting approaches.
There is no universally accepted definition of AI. In general, AI is associated with the development and implementation of computer systems to perform tasks that traditionally would have required human cognitive intelligence, such as thinking and decision making. Machine learning is a subset of AI that generally refers to the ability of a software algorithm to identify patterns and automatically optimize and refine performance from processing large data sets with little or no human intervention or programming. Although AI has existed for many years, interest in applying AI has surged as a result of increases in computing power and the availability of large data sets, including in the financial services sector, “alternative data” not traditionally collected by consumer reporting agencies or used in calculating credit scores. For simplicity, this whitepaper uses the term “AI” to refer to the evaluation of large data sets using machine learning algorithms.
The Current regulatory framework was devised well before AI was available to assist credit underwriting and may constrain the transformative power of AI.
BPI and Covington have prepared this discussion draft to:
- explain the regulatory framework that currently applies to the use of AI in credit underwriting;
- identify the ways in which that regulatory framework may impede broad implementation of AI in credit underwriting; and
- describe how responsible modernization of the regulatory framework would preserve core regulatory principles while removing obstacles to the use of AI to improve credit underwriting.
This discussion draft focuses principally on the regulatory frameworks relating to fair lending and model risk management, as these two areas appear to most impede banks’ efforts to implement AI systems in credit underwriting. In addition, this discussion draft describes the need for a level playing field between banks and non-banks regarding the application of the law to the use of AI in credit underwriting. Modernizing existing regulatory approaches will allow more creditors to utilize AI in credit underwriting, provide consistent consumer protection, strengthen safe and sound underwriting practices, and foster responsible and fair outcomes.
This discussion draft solicits input on identifying and addressing the regulatory obstacles to
implementing AI in credit underwriting. BPI is not presently making policy recommendations. Instead, it is seeking information and analysis relevant to the following potential principles for regulatory modernization in this area:
- The Consumer Financial Protection Bureau (“CFPB”) should lead an effort to modernize the regulatory framework for the use of AI in credit underwriting in light of its authority to implement the nation’s federal consumer financial protection laws and to regulate both banks and non-banks.
- The CFPB should consider, in consultation with the relevant federal agencies:
- developing standards tailored to prevent unlawful discrimination under the Equal Credit Opportunity Act (“ECOA”) in the use of AI credit underwriting systems;
- clarifying that an AI credit underwriting system can qualify as an “empirically derived, demonstrably and statistically sound, credit scoring system,” just like a conventional underwriting system;
- developing adverse action notice standards and adverse action reasons that reflect the broader factors considered in AI credit underwriting systems;
- specifying, in consultation with the relevant federal agencies, the steps that banks and non-banks are expected to take to review AI credit underwriting systems for purposes of compliance with federal consumer
- financial protection laws; and
- ensuring a coordinated approach to the oversight of AI in credit underwriting and the consistent application of the law and regulatory framework to banks and non-banks alike.
This discussion draft is organized as follows:
- Section I describes the promise of AI in improving credit underwriting.
- Section II reviews the current state of the law relating to credit underwriting.
- Section III outlines how responsible modernization of regulatory approaches would facilitate the use of AI in credit underwriting while preserving core regulatory principles.
AI can help banks improve access to consumer credit for underserved consumers in a manner consistent with the fair lending laws and without diminishing banks’ robust credit underwriting standards.
This discussion draft is designed to help identify a path that regulators and the industry can take to achieve these objectives.
You May Also Be Interested In:
