Some lenders are tapping artificial intelligence and machine learning to improve operational efficiency and enhance the borrower experience, but complexities do exist in implementing the technology.
While 63% of lenders are familiar with artificial intelligence and machine learning technology, only 27% have utilized or tested tools for their mortgage company, according to a Fannie Mae report.
Anomaly Detection Automation, which helps detect fraud or defects early-on in the underwriting process, and Borrower Default Risk Assessment, which better predicts the probability of borrower default and supports lenders in taking proactive measures, are among the most appealing AI and machine learning concepts, as cited by lenders.
Though AI and machine learning tech helps reduce the risk of human error and enhances productivity, integrating these applications within a company's infrastructure has proven difficult. Also challenging for lenders are the high costs associated with adoption and the lack of a proven success record for these tools.
Use cases for AI and machine learning tech for the mortgage industry include identifying anomalies, risk assessment, exploring non-credit-bureau data to enhance prediction of loan performance, and answering customer questions through search tools, improved guides and chatbots, according to Fannie Mae.
Mortgage banks are much more likely than depository institutions or credit unions to be familiar with AI and machine learning tech. About 75% of mortgage banks have some form of knowledge about the applications, with 40% having tools already deployed. Only 15% of depository institutions and 10% of credit unions have implemented AI and machine learning tools.
AI and machine learning have gained traction over the past couple years, particularly in industries juggling large amounts of data.
