AI just turned mortgage underwriting into a 47-second process

In partnership with

AI enters one of the slowest corners of finance

Mortgages are notoriously slow.

Anyone who has gone through the process knows the routine. Upload documents. Wait days. Receive requests for more documents. Wait again. Then more verification, more paperwork, and more time.

For decades, mortgage underwriting has been one of the most complex and time consuming processes in financial services. It often takes around three weeks, sometimes longer, to complete the checks required to approve a home loan.

Now one fintech company believes artificial intelligence can compress that process into less than a minute.

Online mortgage platform Better has partnered with OpenAI to launch a new application inside ChatGPT that aims to dramatically accelerate mortgage underwriting. According to the companies, the tool can reduce the time required to approve a mortgage or home equity loan from roughly 21 days to as little as 47 seconds.

If that promise holds up, it could reshape one of the largest financial markets in the United States.

What the new AI mortgage tool actually does

The new ChatGPT application combines Better’s mortgage technology infrastructure with OpenAI’s AI models.

Instead of requiring human loan officers to manually move through dozens of underwriting checks, the system automates many of those steps simultaneously.

Mortgage underwriting involves verifying multiple factors before a lender can approve a loan. These include property appraisals, title reports, borrower income verification, credit reports, and various regulatory checks. Each step typically involves coordination across multiple systems and vendors.

Better’s new platform attempts to handle many of these checks automatically.

According to Better CEO Vishal Garg, the AI system can run parallel workflows across dozens of checkpoints at the same time. This allows loan officers working at banks, mortgage brokers, or fintech lenders to process loan applications far faster than traditional systems.

Instead of reviewing each requirement sequentially, the AI engine evaluates them simultaneously.

The result is a dramatically compressed underwriting timeline.

OpenAI’s chief commercial officer Giancarlo Lionetti said the partnership represents a major shift in how mortgages could be processed in the future.

The goal, according to OpenAI, is to make home financing cheaper, faster, and easier for American families.

The IT strategy every team needs for 2026

2026 will redefine IT as a strategic driver of global growth. Automation, AI-driven support, unified platforms, and zero-trust security are becoming standard, especially for distributed teams. This toolkit helps IT and HR leaders assess readiness, define goals, and build a scalable, audit-ready IT strategy for the year ahead. Learn what’s changing and how to prepare.

Download the Toolkit

A trillion dollar industry full of inefficiencies

The timing of this launch is significant.

The U.S. mortgage market originates more than $1 trillion in home loans every year. Yet despite its size, much of the industry still runs on fragmented systems, legacy workflows, and manual review processes.

Many of the inefficiencies date back to regulatory changes after the 2008 financial crisis.

Following that crisis, large banks such as JPMorgan scaled back their presence in mortgage lending. As a result, a new generation of non-bank mortgage lenders rose to dominance.

Companies like Rocket Mortgage, United Wholesale Mortgage, and Pennymac became major players by building specialized lending operations and focusing heavily on mortgage origination.

While these companies improved parts of the customer experience, underwriting itself has remained slow and operationally expensive.

That is the opportunity AI companies now see.

If artificial intelligence can automate underwriting workflows, it could significantly reduce both time and cost for lenders.

And that is exactly the disruption Better is aiming for.

Better’s pivot: from lender to infrastructure provider

Better originally built its reputation as a direct-to-consumer digital mortgage platform.

But the company is now repositioning itself as something larger.

The new ChatGPT integration reflects Better’s shift toward becoming a mortgage technology platform for the broader industry. Instead of only originating loans for its own customers, the company wants to provide the underlying infrastructure that other lenders can use.

In other words, Better wants to offer mortgage-as-a-service.

Under this model, banks, mortgage brokers, and fintech lenders could plug into Better’s platform to automate underwriting, streamline approvals, and reduce operational costs.

That shift also means the company is taking direct aim at the industry’s biggest players.

Garg has been unusually direct about that goal.

He argues that the dominant mortgage companies generate significant revenue by charging fees tied to underwriting and loan processing. According to his estimate, lenders collectively charge about 1.5 percent of the mortgage value to cover underwriting costs.

Across the industry, that adds up to roughly $20 billion per year paid by American borrowers.

If AI can reduce the cost of underwriting, those fees could shrink dramatically.

That would change the economics of the mortgage industry.

Markets react to the announcement

Investors quickly noticed the potential competitive implications.

Shares of Better jumped as much as 5 percent after the announcement.

Meanwhile, stocks of some major mortgage lenders declined. Rocket Mortgage shares fell by as much as 6 percent, while United Wholesale Mortgage dropped nearly 4 percent during trading.

The reaction highlights how sensitive the mortgage industry is to potential automation.

If AI driven underwriting platforms become widely adopted, lenders with heavy operational costs could face significant pressure.

Why underwriting is a perfect AI target

Mortgage underwriting may look like a niche financial process, but from an AI perspective it is an ideal automation candidate.

The process relies heavily on structured data, standardized verification steps, and rule based decisions.

AI models excel at tasks like reviewing documents, analyzing financial data, and coordinating complex workflows.

Better’s platform uses OpenAI models to process large volumes of mortgage data while navigating what Garg describes as a complex logic tree. The system makes multiple tool calls across different services and maintains a large context window to track all the necessary inputs.

In simpler terms, the AI is acting like a coordinating engine that orchestrates dozens of mortgage checks at once.

That ability to run parallel evaluations is what allows the system to compress weeks of work into seconds.

The bigger fintech implication

Whether Better’s 47 second underwriting claim proves realistic at scale remains to be seen.

Mortgage approvals involve regulatory oversight, risk management, and edge cases that often require human judgment.

However, even partial automation could significantly change the industry.

If AI reduces underwriting times from weeks to hours or minutes, the entire mortgage experience could shift.

Borrowers could receive approvals faster. Lenders could operate with smaller teams. And fintech platforms could build entirely new products on top of automated mortgage infrastructure.

More importantly, this launch signals that AI agents are now moving into core financial workflows, not just customer service or chat interfaces.

From underwriting to compliance to loan servicing, many back office processes across finance remain ripe for automation.

The mortgage market might just be the beginning.

The bottom line

Mortgages have long been one of the slowest processes in financial services.

Better and OpenAI are betting that artificial intelligence can change that.

By embedding mortgage underwriting into a ChatGPT powered system, the companies are attempting to compress weeks of work into seconds.

If successful, the technology could reshape the economics of a trillion dollar industry and challenge some of its largest incumbents.

The race to automate mortgages has officially begun.