Today, around 26 million American consumers are considered credit invisible, meaning they do not have a credit record needed to obtain credit cards, a loan, or a mortgage even though they may be financially responsible.
For Americans who are credit invisible, the use of artificial intelligence (AI) in determining credit worthiness would remove a substantial barrier that prevents them from accessing credit. While AI offers a solution to credit invisibility, evolving regulations around the technology make it unclear to lenders what hurdles must be passed to comply with federal rules. With AI development well underway, federal agencies should now promote a transparent regulatory environment that still encourages innovation around AI.
While traditional lending methods utilize a FICO score to determine creditworthiness, AI can access alternative data to better determine potential borrowers’ credit worthiness. Unlike the FICO score that is based solely on credit history, AI could determine credit worthiness through the examination of a borrower’s deposits and withdrawals, and the applicant’s education and occupation.
The accessibility of this information has brought other forms of credit to many more eligible consumers. Companies such as ZestFinance in Los Angeles have used this technology to offer credit to consumers with little to no credit history who would otherwise be denied credit based on the traditional measures.
For lenders, AI has reduced the number of non-performing loans by 53% while increasing revenue by an average of 37%. The added confidence AI has brought to lenders allows for these companies to better serve existing borrowers as well as reach more eligible consumers.
According to Finextra, a technology news source, AI has reduced the number of unserved consumers seeking loans by 45%. As a result of AI, many more Americans can access loans for big purchases such as a car or a home.
While the benefits of AI are extensive, federal regulations threaten to create unwanted uncertainty for lenders hoping to use AI as part of their application process.
Current laws, including the Fair Credit Reporting Act, prohibit “the use of AI to unfairly deny people employment, housing, credit, insurance, or other benefits.” The Fair Credit Reporting Act does not prevent the use of algorithms, but fears of bias and stiff penalties for violations de-incentivize lenders from using AI.
To incentivize lenders to adopt AI, regulators have to promote policies that allow for this innovative technology to be used while also protecting consumers from potential biases and privacy breaches in algorithms.
The implementation of Trial Disclosure Programs (TDP) policy by the Consumer Financial Protection Bureau (CFPB) is one example of a policy that has encouraged innovation while also protecting consumers. TDP allows for in-market testing of AI at the permission of the CFPB, providing a safe harbor for lenders who want to deploy this technology. By protecting lenders from potential penalties, these programs have removed the uncertainty associated with AI adoption and helped identify weak spots in the algorithms of participating lenders, leading to better services for consumers.
In order to encourage the use of AI, the Federal Trade Commission and other government agencies should explore new ways to protect consumers while also encouraging innovation in the credit market. Using distributed ledger technologies, a form of blockchain technology, would increase data transparency through the permanent storage of transaction records. These blockchains securely store consumer data in “blocks” and link different types of data together in “chains.” The utilization of blockchains with AI in credit markets would guarantee consumer data remains protected while also promoting innovation.
AI has the power to offer credit to more consumers through the use of alternative data that provides a more holistic view of a potential borrower. While the future of lending with AI is bright, federal regulators must work to alleviate the uncertainty current regulations have created. Not doing so could mean lost innovation and fewer Americans having access to credit.