Google Cloud’s Lending DocAI Tackles Mortgage Paperwork

The artificial intelligence offering, designed to reduce the time and cost of closing home loans, streamlines data capture from the reams of scanned and digital documents detailing borrowers’ income, assets and other application information.


Google Cloud wants to speed up mortgage processing with the first industry-specific offering leveraging its machine learning-powered Document AI technology introduced last year.

The new Lending DocAI is designed to reduce the time and cost of closing loans for the mortgage industry. The artificial intelligence (AI) offering streamlines data capture from the reams of scanned and digital documents detailing borrowers’ income, assets and other loan application information.

Lending institutions process hundreds of pages of paperwork for each loan they handle, a largely manual, time-consuming task that can add thousands of dollars to the cost of issuing a loan, according to Sudheera Vanguri, Google Cloud’s Document AI product manager.

Sponsored post

“Unlike more generalized competitive offerings, Lending DocAI provides industry-leading data accuracy for documents relevant to lending,” Vanguri wrote in a blog post today. “Lending DocAI leverages a set of specialized models focused on document types used in mortgage lending and automates many of the routine document reviews, so that mortgage providers can focus on the more value-added decisions.”

Lending DocAI can parse text, characters and images in documents ranging from a content-heavy invoice with tabular data to a camera-captured image of a wrinkled and wine-stained W-2 U.S. tax form with skewed text, according to a video presentation scheduled for the Mortgage Bankers Association’s (MBA) annual convention that starts Monday. Now in preview, the offering is expected to be generally available next year.

In addition to helping process loan applications more easily, Lending DocAI helps organizations meet regulatory and compliance requirements, according to Google Cloud. Organizations can reduce risk and enhance their compliance postures by “leveraging a technology stack—e.g., data access controls and transparency, data residency, customer-managed encryption keys—that reduces the risk of implementing an AI strategy,” Vanguri said. “It also streamlines data capture in key mortgage processes such as document verification and underwriting.”

Google Cloud will share information about a Lending DocAI partnership with San Francisco-based Roostify at the online MBA convention Monday. The companies have been working on a joint offering that uses Lending DocAI to power new and enhanced document capabilities in Roostify’s digital mortgage lending platform. The new service will be in production in early 2021, according to a Google Cloud spokesperson. The same functionality will be offered in a stand-alone Document AI With Roostify service that lenders can plug into their existing systems.

“The mortgage industry is still early in transitioning from traditional, manual processes to digitally enabled and automated, and we believe that transformation will happen much more quickly with the power of AI,” Roostify founder and CEO Rajesh Bhat said. “And if you are going to do AI, you’ve got to go Google.”

Google Cloud will be adding other platform integration partners for Lending AI, which its channel partners can resell and service.

Under CEO Thomas Kurian, Google Cloud has adopted a vertical, industry-specific solution strategy to help win enterprise customers in six target industries: financial services, health care, manufacturing and industrial, the public sector, retail, and media, telecommunications and entertainment.

It’s also working on Procurement DocAI, which leverages its Document AI technology to help organizations automate their procurement cycles, according to a spokesperson.

“Google Cloud provides a group of AI-powered parsers … that take documents [such as invoices and receipts] in a variety of formats and return cleanly structured data,” the Google Cloud spokesperson said. “We’re working on solving important use cases for other high-volume document workflows in industries like manufacturing, health care, financial services industries and more.”