When Robots Process Your Home Loan
AI’s Role in the Mortgage Industry
Artificial Intelligence (AI) is no longer a futuristic, sci-fi concept. In fact, it’s likely you’re already interacting with AI every day without realizing it. You use “supervised AI” when you ask Siri to provide driving directions. It’s supervised because your human prompt sets off an automated process.
If you own a Roomba or a self-parking car, AI is also doing something called “unsupervised AI”. Your Roomba is quite literally learning the measurements of your home and then applying that knowledge to alter its behavior. It’s making decisions and taking action, without any prompt from you. The AI in your self-parking car takes things a step further by using machine learning to differentiate between a car, person or tree when backing into a space. Pretty wild.
AI is also playing a role in your homeownership journey, especially when it comes to your loan approval process. Housing Wire predicts that by the end of 2020 most mortgage applications will involve unsupervised AI, with little to no human interaction. From beginning to end, your entire loan approval process could be handled entirely by AI.
So what does the increasing use of technology in loan processing mean for home loan applicants? If all goes as predicted, it means faster processing and more comprehensive approval criteria. Below are two big examples.
The Forbes article “AI Is Coming To Take Your Mortgage Woes Away” discusses how labor-intensive applying for a home loan can be for prospective buyers. Applicants need to gather “two years’ worth of tax returns, W-2s, bank statements, copies of identification plus any name changes, proof of other income, and so on.”
On the flip side, lenders need to organize input and evaluate all the information you dug out of file cabinets in order to determine if you qualify. AI was built to take on the burden of labor-intensive data, doing it faster and more accurately than your loan processor could. “For instance, while right now you could expect to wait about three weeks for a mortgage application to be approved, AI will probably reduce that to one day within a few years.”
Because AI is able to capture and analyze massive amounts of data by itself – it can expand the parameters of your loan approval process. Historically, your home loan approval has been heavily based on your three-digit FICO score. The problem is that FICO is just one part of the picture when it comes to determining creditworthiness. Forbes writes that, “Mortgage applicants are complex people whose ability to pay their home mortgage is best foretold by looking at the whole mosaic of their lives.”
Making FICO the primary decision-maker when it comes to mortgage approval can also discriminate against minority applicants. Matt Diffanis, past President of Illinois REALTORS® and fair housing advocate explains, “FICO score minimums tend to disadvantage those who don’t use credit (e.g., immigrants who use cash or borrow from family and friends, leaving no credit footprint) or a black renter whose FICO score is low due to medical collection, but who has a spotless rental and utility payment history.”
The National Association of REALTORS® (NAR) agrees that homeownership shouldn’t be out of reach for low-income, rural and minority borrowers who lack access to traditional forms of credit. NAR supports ongoing efforts to take into account consumer payment history not typically considered, such as rent, utility, telephone, and other regular payments and urges regulators, the Consumer Financial Protection Bureau (CFPB), the Government-Sponsored Enterprises (GSEs), and lenders to work to strengthen these efforts.
Presently, there is not an across-the-board standard that lays out what the next level of data AI should be looking at during the approval process. It’s a bit of a wild west right now, with startups like LenddoEFL and Zestfinance creating algorithms that explore very new ways to determine creditworthiness.
MarketWatch’s article “How artificial intelligence could replace credit scores and reshape how we get loans” explains how in addition to your financial data, Lenddo NFL’s AI also looks at your online and smartphone behavior.
“Whether you avoid one-word subject lines (meaning you care about details) and regularly use financial apps on your smartphone (meaning you take your finances seriously). Lenddo also looks at the ratio of smartphone photos in your library that were taken with a front-facing camera, since selfies indicate youth, helping the company divide people into customer segments.”
Up until now most of the AI that the banking industry has relied on has primarily been supervised AI. The mortgage processor inputs the data that they have determined is the most important. And a human-dictated process called an algorithm is created, instructing the AI on how to process the data that has been entered. AI is rapidly speeding up the number crunching to shorten the time it takes to get a mortgage approved – but overall, it’s not making decisions by itself.
However, with AI’s capacity to objectively look at more data than we could ever imagine getting our heads around, unsupervised AI is coming soon. Your home loan may be approved without any human interaction at all, making the process even faster and more objective. The vast amount of data that AI can access will likely also influence which parameters the banking industry uses to determine your creditworthiness in the future.