Industry experts detail the benefits, challenges of AI

As artificial intelligence continues to grow in prominence, mortgage professionals “must carefully evaluate and manage” their use of AI and “focus on deriving the benefits while avoiding potentially catastrophic risks.”

That’s one of the key conclusions reached by the BlackFin Group in a recently published white paper, “Artificial Intelligence (AI) in Mortgage Banking.” The paper was co-authored by several BlackFin executives and mortgage technology leaders from other organizations, including Chuck Iverson at Mason-McDuffie Mortgage and Maria Moskver at Cloudvirga.

The paper notes that “AI is not a homogenous technology” and offers a variety of uses across the mortgage ecosystem, from origination and servicing to default solutions and asset sales. The authors outline six of the most common types of AI — machine learning, deep learning, natural language processing, generative AI, expert systems and cognitive computing — while excluding two others (rule-based systems and robotic process automation) that are less relevant to their definition of AI.

They argue that understanding the technology is imperative when choosing a specific tool to deploy.

“In our view, what distinguishes AI is the ability to address situations that are not precisely like the ones it has previously addressed,” the authors state.

Data from Precedence Research shows that the size of the global AI market is estimated to grow from $454 billion in 2022 to $2.5 trillion in 2032. But even with this expected influx of investment capital and user demand, BlackFin’s paper finds that mortgage companies are struggling today to implement AI tools in an efficient manner.

The authors cite some examples, including automated document processing and underwriting systems, in which a company’s expenses have increased but productivity hasn’t.

“Lenders frequently comment on the lack of ROI on technology as costs have risen, even if much of that increase can be attributed to an increase in sales compensation,” the authors write.

They go on to describe the potential significance of AI in multiple areas of mortgage lending. It can reduce the costs to manufacture or service a loan. It can accomplish tasks that humans or other types of technology cannot. And it can fundamentally change the origination and servicing processes. But the authors also stress that none of this should be expected to happen quickly.

“There is little evidence so far that AI can fundamentally transform our industry in the next 5-10 years — there are too many structural and regulatory impediments for that to be the case,” they wrote.

“I think if you want to innovate, you need to be able to think long-term. I don’t think anyone’s ever innovated in the short-term,” Rechat CEO Shayan Hamidi recently told HousingWire. “So you need to be able to have the appetite for that: be willing to take the risks and be willing to be patient for quite some time. And I think AI is one of those things. You can do some fun, cool stuff with it very quickly, but then if you want to start doing meaningful things, it’s a big long-term investment, at least today.”

BlackFin Group — founded in 2019 and based in Englewood, Colorado — is a management consulting firm that helps to guide strategic decisions and find innovative solutions for banks, nonbanks and credit unions across the country. In 2022, it launched a practice dedicated to reverse mortgages.