Skip to content

As Mortgage Fraud Spikes, Lenders Turn To AI For Help

The latest advances in technology have led to revolutionary changes in almost all parts of modern life. The rise of the internet, in particular, has enabled an uncountable number of new services, many of which have made our lives immeasurably easier. One of the most important examples of this is the global shift to online financial services, from banking to investments to lending. On the whole, that’s been a positive development for consumers around the world. It hasn’t come without a considerable cost, though.

In addition to making our financial lives easier, the internet has also given rise to whole new varieties and methods of fraudulent behavior. It has also made the world of fraud borderless, connecting willing participants to schemes located in other parts of the world. One glaring example of this change can be seen in the rising rate of fraud in the mortgage industry. Fortunately, the tech sector is developing an answer to the problem in the form of fraud detection AI systems. Here’s a look at how technology is enabling fraud in the mortgage industry and how technology could one day stamp it out completely.

A Pronounced Spike

In the United States, financial analytics firm CoreLogic has reported that the risk of fraud in the mortgage market has spiked by 12.4% year-over-year in the second quarter of 2018. The data was alarming enough that mortgage giant Fannie Mae has issued a warning to its affiliated lenders about the increasing risks, including a glimpse of what may be fueling it. At issue, they say, is that rising interest rates and skyrocketing home prices are driving many applicants to overstate their earnings in an attempt to qualify for loans that are beyond their means. Ordinarily, lenders cross-check and verify income statements from applicants, and in the past, detecting inflated earning statements was a simple matter of following up with employers. Now, the internet is providing a way to fool even seasoned fraud prevention specialists.

Would You Like Pay Stubs With That?

On the internet, there’s never a shortage of unscrupulous actors looking to earn money, no matter the legality of their methods. That has translated to a variety of services operating openly online that offer mortgage applicants ways to fool lenders with inflated earnings statements. The services, which are easy to find, will not only provide falsified pay stubs and financial documentation but also offer phone verification of the information they contain. That means that mortgage companies looking to verify application data will be duped into calling a false phone number, where a real live operator (operating out of overseas call centers) will vouch for the documents in question. The Federal Trade Commission has taken action against several such outfits, but like a game of whack-a-mole, more spring up to replace them.

AI To the Rescue

For its part, the mortgage industry seems to be realizing that they need to take new steps to curb the rising rate of fraud. To tackle the problem, they’re turning to a whole new generation of AI-powered analytics systems that have the capability to comb through vast amounts of data to find patterns indicative of fraud that would otherwise go undetected. A McKinsey survey indicated that a majority of risk managers believed that the new tools would reduce credit decision times by 25 to 50 percent, and reduce losses due to fraud by at least 10%. That’s a good start, but companies like ID Finance think they can do far better. Their platform, aimed at emerging markets (where documentation standards are lower and fraud is rife) has already seen success in rooting out fraudulent applications early on in the loan process. Their system analyzes previous fraudulent applications using machine learning techniques to develop smarter filters. That means that each fraudulent application that makes it through the system provides more data to make the system even smarter going forward.

The Future of Smart Lending

Right now, the US mortgage market is serving as something of a test bed for the latest in AI anti-fraud technology, but the results will have implications that reach around the globe. That’s because a look at any mortgage market comparison tool in other major real estate markets reveals conditions that mirror what’s been happening in the US. As prices continue to rise, and easy credit becomes a thing of the past as central banks raise interest rates to fight inflation – it’s safe to assume that we’ll see similar increases in mortgage fraud rates start to show up around the world. That’s going to make the latest in AI anti-fraud solutions all the more important, and the early results offer plenty of hope for the future. In short, if you’re thinking about fudging some numbers on an upcoming mortgage application, you might want to think again – an AI might already be watching.


Up Next