Adrian JohnsonTLDR:
– Consumers in the US are increasingly being rejected for loans, memberships, and job opportunities by AI systems, without any recourse or appeal.
– The use of AI in decision-making processes is amplifying existing biases and discriminating against certain demographics.
– Lack of credit history, collateral, and adherence to traditional metrics of financial responsibility are limiting social mobility, particularly for younger generations.
– The Consumer Financial Protection Bureau has mandated that credit lending agencies explain the reasoning behind loan denials.
– Experts argue for the consideration of alternative data, such as rent and utility payments, social media behavior, and spending patterns, in loan decision-making.
Consumers in the US are facing increasing rejection for loans, memberships, and job opportunities as companies turn to artificial intelligence (AI) systems to make key decisions. The use of AI in these processes is amplifying existing biases and discriminating against certain demographics. For instance, black Americans are 80 percent more likely to be auto-rejected by loan granting agencies than their white counterparts, according to a 2021 investigation. The biases are perpetuated by training AI models on historical data, which reflects longstanding discriminatory practices in the banking industry. Companies have even been accused of granting lower credit limits to women compared to men. This denial of opportunities also limits social mobility for younger generations, who may lack credit history and collateral. The use of traditional metrics, such as credit cards, homes, and cars, to assess financial responsibility is becoming less relevant in today’s society. Only about half of Gen Z individuals have credit cards, and fewer are opting to obtain driving licenses, which limits their collateral when applying for loans.
The Consumer Financial Protection Bureau (CFPB) has implemented safeguards to require credit lending agencies to explain the reasoning behind loan denials. However, consumers often face challenges in appealing these decisions due to the lack of a human appeal process. As a result, some individuals, like the subject referred to as “D” in the article, have had to postpone bill payments and worry about their long-term financial health. Experts argue that lenders should consider alternative data, such as rent and utility payments, social media behavior, and spending patterns, in loan decision-making. This wider range of data points could provide a more accurate representation of an individual’s financial responsibility. However, the increasing reliance on AI-driven tools, such as applicant tracking systems (ATS) and chatbots, in job applicant evaluations is also causing concern. These systems often overlook transferable skills and non-linear career paths, excluding qualified applicants, particularly those from diverse backgrounds.
