What the $450 million Da Vinci auction means for credit application fraud

Just as an art buyer parting with almost half a billion dollars must be confident that the work they buy is genuine, and not from a disciple, banks need to be sure that a credit application has been made by a real person and not a fraudster. Credit application fraud is a big problem for banks, and is only getting bigger with the advent of digital loan origination channels. To try and detect fraud, banks give scores to credit applications in order to estimate the likelihood of it being fraudulent. How much better could those estimations be if banks adopted "deep learning"? The answer is "a lot better".