Artificial intelligence is unlikely to upend the mortgage industry, but it could make the biggest lenders even stronger by lowering costs, speeding up loan production and accelerating industry consolidation, according to a July 12 report from investment bank Keefe, Bruyette & Woods (KBW).

KBW pushes back against the growing narrative that AI will broadly disrupt financial services. Instead, the firm argues that the industry’s largest players are best positioned to benefit because they have the scale, customer data, regulatory infrastructure and capital needed to deploy AI effectively.

The report looks at several sectors in the financial services industry, including exchanges, consumer finance, traditional banks, mortgage banking and title insurance, among others.

KBW’s broader analysis found that financial sectors with significant regulatory oversight, proprietary data and large technology budgets are expected to benefit the most from AI adoption.

The firm ranked exchanges as the sector least vulnerable to AI disruption, followed by consumer finance and the nation’s largest banks, arguing that these businesses are more likely to use AI to improve efficiency than face displacement.

Mortgage banking fell closer to the middle of KBW’s risk analysis. The company said the industry’s repetitive, rules-based workflows make it well suited for automation, but they expect AI to reinforce the advantages of large lenders rather than fundamentally reshape the business.

Mortgage insurance companies and mortgage real estate investment trusts (REITs) were ranked as carrying somewhat higher AI disruption risk, although KBW said these businesses also benefit from regulatory protections and capital-intensive business models that limit the threat of displacement.

Mortgage banking outlook

In mortgage banking, KBW analysts see AI as a productivity tool rather than a disruptive force, citing that several steps in the mortgage process remain labor-intensive and repetitive. Automating these tasks could shorten loan cycle times, reduce costs and improve accuracy, the report said.

KBW ranked mortgage banking among the financial sectors facing relatively high competitive pressure from AI, assigning it a risk score of 4.65 on a 10-point scale.

Analysts said the greater risk is not that AI replaces mortgage lenders; it’s that larger companies will pull further ahead while smaller competitors struggle to keep up with technology investments. The report points to workflow automation, improved servicing economics, correspondent lending pressure and faster industry consolidation as the biggest trends to watch.

Servicing could be one of the biggest beneficiaries of AI adoption, KBW analysts noted. Because servicing involves large volumes of repetitive, data-driven work, AI could lower servicing costs while helping lenders better identify refinance opportunities and retain existing borrowers.

Among mortgage companies, KBW highlighted Rocket Mortgage as one of the firms best positioned to benefit.

“As one of the largest IMBs, Rocket has the scale, proprietary borrower data, servicing portfolio, and capital to operationalize AI, and we believe disruption across the industry should accrue to the biggest platforms rather than threaten them,” the report said.

Analysts added that they expect Rocket’s investments in AI and machine learning to improve borrower retention, increase refinance recapture rates, and reduce expenses across the origination and servicing channels.

More broadly, KBW expects AI to widen the gap between industry leaders and laggards rather than transform mortgage lending itself.

KBW said it is cautious about smaller regional and community banks because AI is “likely to become ‘table stakes’ rather than a source of durable differentiation.

“These institutions generally lack the technology budgets, internal AI talent, and structured proprietary data needed to build differentiated AI capabilities in-house, leaving them more dependent on third-party vendors and core providers,” the report said.

This article was written by Sarah Wolak and generated with the assistance of HousingWire Automation, then reviewed by a HousingWire editor before publication.

More: