CFPB reiterates skepticism of AI in fair-lending report

The Consumer Financial Protection Bureau reiterated its skepticism of machine learning and predictive analytics in a fair-lending report issued Friday that shows an increase in actions against redlining and discriminatory practices.

The CFPB said financial services will be increasingly shaped by predictive analytics, algorithms, and machine learning but that technology can also reinforce “historical biases that have excluded too many Americans from opportunities.”  

In the report, Patrice Ficklin, the CFPB’s fair-lending director, said that while she is encouraged by programs that can expand access to credit, she is skeptical of claims that advanced algorithms are a “cure-all” that can eliminate bias in credit underwriting and pricing. Her boss, CFPB Director Rohit Chopra, had previously warned companies about relying too heavily on AI and machine learning in making lending decisions. 

Ficklin also noted a “legacy of structural discrimination” that extends to mortgage, auto, student and other credit markets for people of color, in the report.

“The mission of fair lending is to break these discriminatory patterns and practices and to promote access to credit to create fairer markets for all,” she said.

Last year the CFPB issued four fair-lending actions alleging violations of the Equal Credit Opportunity Act and referred two matters involving allegations of discrimination to the Department of Justice, according to the 43-page report. 

In 2020, under former acting CFPB Director Dave Uejio, the agency filed one lawsuit and issued four Justice Department referrals. 

The CFPB said it will be “sharpening its focus on digital redlining and algorithmic bias.” As more Big Tech firms offer financial services and products, the agency said it will be working to identify emerging risks and to develop appropriate policy responses.

Ficklin also said the CFPB “will continue to fight discrimination that manifests as unfair, deceptive, or abusive acts and practices,” known as UDAAP violations. 

“Redlining, pricing discrimination, and appraisal bias are significant barriers to fair competition in the mortgage market, impeding the ability of an individual borrower to get credit on fair terms, thus stifling growth in communities across the country,” Ficklin wrote.