The CFPB acts to protect the public from black box credit models using complex algorithms

      Comments Off on The CFPB acts to protect the public from black box credit models using complex algorithms

washington d.c. – Today, the Consumer Financial Protection Bureau (CFPB) confirmed that federal anti-discrimination law requires businesses to explain to applicants the specific reasons for denying a credit application or taking other adverse action, even if the creditor relies on credit models using algorithms. The CFPB published a Circular on consumer financial protection to remind the public, including those responsible for enforcing the Federal Consumer Financial Protection Act, of creditors’ notice of adverse action requirements under the Equal Opportunity Act in credit (ECOA).

“Companies are not exempt from their legal responsibilities when they let a black box model make lending decisions,” CFPB Director Rohit Chopra said. “The law gives every applicant the right to an accurate explanation if their credit application has been denied, and that right is not diminished simply because a company uses a complex algorithm that it does not understand.”

The collection of data about Americans has become large and ubiquitous, giving companies the ability to know very detailed information about their customers even before interacting with them. Many companies across the economy rely on these detailed datasets to power their algorithmic decision-making, which is sometimes marketed as “artificial intelligence”. The information gathered from data analysis has a wide range of commercial uses by financial companies, including for targeted advertising and credit decision making.

Law-abiding financial companies have long used advanced calculation methods as part of their credit decision-making processes, and they have been able to provide rationales for their credit decisions. However, some creditors may make credit decisions based on the results of complex algorithms, sometimes referred to as “black box” models. The reasoning behind some of the outputs of these models may be unknown to model users, including model developers. With such models, adverse action notices that meet ECOA requirements may not be possible.

ECOA protects individuals and businesses against discrimination when seeking, applying for, and using credit. To help ensure that a creditor does not discriminate, the ECOA requires that a creditor provide notice when taking adverse action against a plaintiff, which must contain specific and specific reasons for such adverse action. Creditors cannot legally use technologies in their decision-making processes if their use means that they are unable to provide the required explanations.

Today Circular specifies that :

  • Federal consumer financial protection laws and adverse action requirements must be applied regardless of the technology used by creditors. For example, the ECOA does not allow creditors to use technology that prevents them from providing specific and specific reasons for adverse actions. The use of complex algorithms by creditors should not limit the application of the ECOA or other federal consumer financial protection laws.
  • Creditors cannot justify non-compliance with the ECOA on the basis of the simple fact that the technology they use to assess credit applications is too complicated, too opaque in its decision-making, or too new. Creditors who use complex algorithms, including artificial intelligence or machine learning technologies, to make credit decisions must always provide a notice that discloses the specific primary reasons for taking adverse action. There is no exception for breach of law because a creditor uses technology that has not been properly designed, tested or understood.

Whistleblowers play a pivotal role in uncovering information about companies using technologies, like black box models, in ways that violate ECOA and other federal consumer financial protection laws. Having clear and actionable information is essential for the CFPB and other consumer protection officials. The CFPB encourages technicians to provide information to the agency, and they can visit the CFPB Whistleblower Program webpage to learn more.

In addition to whistleblowers, government partners are also critical to CFPB’s law enforcement efforts. For example, the CFPB closely follows the work of the National Institute of Standards and Technology within the US Department of Commerce and other government agencies around the world, to assess the benefits and risks associated with emerging technologies.

The risks associated with decision-making technologies extend beyond adverse action notices and the ECOA. Recently, the CFPB has begun to look closely at the use of automated valuation models as part of the home appraisal process to ensure that home appraisals are accurate and fair.

Read today’s consumer financial protection circular, Adverse Action Notification Requirements for Credit Decisions Based on Complex Algorithms.

Consumers can file complaints about Fair Lending or complaints about financial products or services by visiting the CFPB website or calling (855) 411-CFPB (2372).

###

The Consumer Financial Protection Bureau is a 21st century agency that implements and enforces federal consumer finance law and ensures that markets for consumer financial products are fair, transparent and competitive. For more information, visit consumerfinance.gov.