It’s only fair. Predictive analytics is a win/win for both borrowers and lenders.


New CFPB study shows AI and machine learning can approve significantly more applications, while yielding lower average APRs; AI Lift proves itself twice as predictive as the competition

It’s a risk-based business facing rapid changes and a world of unknowns. But surprisingly, the credit industry is now poised to be more equitable, efficient, and predictable than ever. Enlisting advanced analytic tools including AI and alternative data, credit risk web services like AI Lift are delivering a new standard of access and fairness. Fairness for the millions of creditworthy applicants unable to access lower interest rates through traditional screening methods, and fairness for lenders who need a frictionless method for reaching the “invisible” customers who can grow their business.

The advantages of predictive analytics was recently confirmed by the Consumer Financial Protection Bureau (CFPB). In a study of underwriting models, the CFPB concluded that methodology using alternative data and machine learning approved 27% more applications than a traditional lending model and yielded 16% lower average APRs. Additionally, the expansion of credit access reflected across all tested race, ethnicity, and sex segments and significantly expanded access in many consumer segments, such as “near prime” consumers, applicants under 25 years of age, and consumers with incomes under $50,000.  

But not all screening methods are created equal

In a recent data test of 18 competitive credit screening scores, AI Lift from Accelitas delivered twice the predictive lift over the vendor average — at 77% the cost of our nearest competitor. By identifying more good customers, AI Lift’s advanced analytics can deliver a ROI of 30:1, creating a win/win scenario for everyone.

Speaking of Fair….

To ensure compliance with FCRA regulations, it’s important that any “no” decision is explainable, even if the decision to turn down a consumer seeking credit was made with the assistance of AI-powered analytics. AI Lift delivers insights that are both predictive and interpretable. Our Explainable AI enables lenders to apply AI techniques at the top of a data waterfall, accepting more profitable accounts before additional data costs are incurred, while providing the information necessary to support lender compliance with FCRA adverse action notice requirements.

We’ll dig deeper into that in a future blog

Want to learn more?
Download our 


Tags: Machine Learning, lending, AI Lift, Accelitas, Artificial Intelligence, linear model, interpretable results, Credit Risk Web Service, Credit Risk, Adverse Actions, Explainable AI, predictive analytics, CFPB, near prime, FCRA

Posted by Greg Cote on 9/18/19 7:00 AM