Mobile Finance Sure Is Convenient; Fraudsters Think So, Too

We love the speed and convenience of making auto loan payments and conducting other household business on mobile devices.

But those same qualities also give lenders less time — trending, literally, to zero time — to catch fraudsters by the time a payment transaction is executed and money changes hands electronically.

Artificial intelligence is increasingly how cybersecurity experts weed out phony transactions before they start, or flag a phony applicant after credit is extended, said Doug Clare, vice president of security solutions and FICO analytic cloud for Fair Isaac Corp., better known as FICO.

“Increasingly, you have zero time to clear a transaction,” Clare said in a phone interview. “The time between initiation and conclusion is being driven down to zero.”

The auto finance industry in particular has a growing issue with so-called synthetic identity theft. That’s where fraudsters secure credit cards and other credit, using a mix of fake and genuine consumer data. They mimic an honest consumer and pay bills on time, long enough to add more credit cards, and maybe even an auto loan, before cashing out and disappearing.

In opposition to that, artificial intelligence learns from experience, and cross-references mind-boggling amounts of data. AI can’t slow down the clock, but it can help create systems that are radically quicker to recognize a transaction that smells likely to be fraudulent. Even if the details of a specific transaction appear to be OK, for instance, a transaction could fit a suspicious pattern that’s less readily apparent without AI.

Clare said that AI adds a new layer of “less manipulate-able” criteria, on top of lower-tech criteria lenders already use to authenticate that an applicant is who they appear to be.

“Fraudsters are really good at figuring out how to answer the questions to get what they want,” he said. “AI is harder to reverse-engineer.” One thing for sure, Clare said: “The fraudsters aren’t going to quit trying.”