At Blend Forum 2025, our annual executive gathering where more than a hundred leaders from the nation’s top banks, credit unions and IMBs gathered to discuss the future of lending, one theme stood out. In his opening remarks, Nima Ghamsari, co-founder and head of Blend, put it clearly: the speed of technology adoption is now the decisive advantage for lenders.
Institutions that move quickly from pilot to practice, with use cases that reduce costs and strengthen relationships, will set the pace for the industry. The urgency of that message was palpable in every discussion. Lenders do not need more additional tools that increase complexity. They need intelligent systems that are at the heart of creation and actually do the work.
Beyond digitalization
The sector has invested heavily in digitalization over the past ten years. Online applications, e-signatures, borrower portals and automated verifications have transformed the front end of the borrower experience. These investments paid off in higher throughput rates, shorter cycle times and better engagement.
But the core economy has not changed. Origination still costs $10 to 12,000 per loan, cycle times average 20 to 30 days, and exceptions still return files to human hands. What digitalization modernized were the touchpoints, not the process itself. Files are held in queues, documents are checked manually and quality control takes place afterwards.
The result is an industry burdened by rigid workflows as consumer expectations and market pressures increase.
Why agentic AI is a step change
The opportunities for AI in mortgage lending are not just about accelerating existing steps. It’s about rethinking how the process works as a whole. Traditional rules-based automation can pass a file, but this happens in the gray areas where most lending actually happens.
Agentic AI changes that equation. These systems interpret information, reconcile inconsistencies, and act independently while knowing when to call in a human for supervision. Documents are not only digitized, they are also understood. Conditions are not only flagged, they are also resolved. Creation is less about the processing and more about managing the results.
This represents a true step change: from static workflows to dynamic, continuously executing systems. It’s the difference between an assembly line that stops when something doesn’t fit, and a system that adapts on the fly to keep production going.
Early pilots point to what’s next
The shift from theory to practice is already underway. Forward-thinking lenders are testing agentic AI capabilities that go beyond surface-level automation and into the execution layer of origination. Blend is testing applied use cases within its platform to demonstrate how AI can handle more of the heavy lifting across the lifecycle.
Document intelligence now classifies and verifies files in seconds, extracting crucial data and flagging discrepancies that once required hours of review. Conversational intelligence helps loan officers by synthesizing conversations, surfacing intent signals, and providing real-time coaching that strengthens both compliance and conversion.
Another promising area is quality control. Manual audits of hundreds of documents and thousands of controls have long been a drag on productivity and a source of costly risk. Early pilots show that AI can perform this assessment dynamically and deliver a transparent quality score in minutes. The result is not only efficiency, but also stronger credit quality and greater investor confidence.
Together, these pilots illustrate what the next chapter of origination could look like: a system where AI is not an afterthought, but an active participant in advancing loans.
The competitive imperative
These examples show what’s possible, but also highlight a widening gap between experimentation and business value. According to recent surveys, 80% of institutions are experimenting with AI, but less than 5% have taken these efforts into production. Too many initiatives remain isolated, disconnected from workflows and ultimately fail to deliver measurable results.
That tension was clearly evident during our AI Roundtable. Some lenders are just beginning to test AI in limited use cases, such as document review. Others are testing broader applications, such as internal copilots or knowledge repositories, but are struggling to scale governance, data quality and adoption across the enterprise. In many cases, individual employees are experimenting faster than corporate programs can keep up, creating a patchwork of adoption levels within the same organization.
For lenders, the challenge is no longer whether AI works in theory. It’s about moving from distributed pilots to systems that have a material impact on costs, certainty and growth. Institutions that embed intelligence at the core of origination, rather than trapping it at the edges, will make progress in both efficiency and borrower experience
Looking ahead
The future of lending belongs to those who adopt systems that not only digitize processes, but also actually think and act independently. By going beyond experimentation and embedding intelligence at the execution layer, lenders can create a fundamentally different business model – one defined by speed, certainty and trust.
For more information about Blend
Related
#simple #automation #embedded #intelligence #future #mortgage #lending