David shares how AI is transforming commercial lending by streamlining workflows, boosting accuracy, and letting bankers focus on stronger client relationships.
David, you’ve had a remarkable career as a serial entrepreneur, building businesses across financial services, e-commerce, and energy. Could you share a bit about your professional journey and what has shaped your approach to combining business and technology?
Relationships have always been at the center of how I think about building companies. Banks and the technology firms that support them have to trust and learn from each other. Given the nature of the industry, it’s incredibly difficult for a community bank to build complex software internally, just as it’s not feasible for a fintech to become a bank.
That reality has shaped how I approach starting a business and setting strategy. I spend a lot of time listening to bankers, understanding where their operations slow down and identifying where technology can provide meaningful support without disrupting what already works.
In the case of commercial lending, bankers consistently told us that one of their most important and risk-sensitive lines of business was still heavily dependent on spreadsheets and manual work. In our research, we found that roughly 77% of U.S. commercial lenders still rely on manual processes. Conversations with lenders revealed that the lending process itself is a competitive advantage. Deal structures vary widely, underwriting approaches differ and institutions don’t want technology forcing them into a one-size-fits-all model.
That insight led us to focus on removing manual labor while preserving lender control. We partnered closely with banks to understand where processes align and where they differ and built software that supports those differences rather than erasing them. That collaboration continues to shape our solutions and ultimately helps our clients operate more efficiently, manage risk more effectively, and stay competitive.
What are the biggest bottlenecks in today’s commercial lending process, and how can AI help solve them?
The biggest bottlenecks in commercial lending happen leading up to the credit decision. Lenders spend a significant amount of time entering figures from tax returns and financial statements, maintaining spreadsheets and assembling credit memos before they can even begin real credit analysis. For many institutions, especially those operating with lean teams, preparation work quickly becomes the limiting factor on both speed and capacity.
Beyond delays, this approach creates structural vulnerability. Small data entry mistakes, formula changes in spreadsheets or inconsistencies between documents can materially affect how a loan is evaluated. When technology is used to automatically read financial documents and apply consistent underwriting calculations, it allows lenders to focus their time more effectively, remove risk and give a more reliable foundation for their decisions.
How are banks moving away from manual, Excel-based workflows with the help of AI?
Spreadsheets persisted in commercial lending because institutions needed flexibility and control over how deals were underwritten. Each lender developed their own approach, and spreadsheets were the only tools that could keep up.
What’s changing now is where that flexibility lives. AI allows banks to bring spreading, document review and data management together into systems that are built specifically for commercial lending. Instead of maintaining dozens of offline files, lenders can work from structured data that flows through the credit process consistently. The result is fewer handoffs, less rework and far more confidence in the numbers being used to make decisions.
Vine’s platform can complete in one hour what traditionally takes a week—how does AI make this level of speed possible?
In commercial lending, or any type of lending, a quick turnaround only has value if the outcome can be trusted. As mentioned earlier, extended timelines aren’t driven by analysis itself, they’re usually driven by the volume of groundwork that has to be laid beforehand.
Credit analysts can spend over a week spreading financials across related entities and reworking documents whenever assumptions change. Automating document reading and applying standardized calculations removes those delays entirely. Instead of waiting on file assembly, lenders can make accurate and confident decisions around the loan. The outcome is a process that moves rapidly without sacrificing confidence in the numbers behind the decision.
In what ways does AI improve the accuracy and consistency of credit decisions?
Accurate credit decisions rely on stable inputs. When figures are entered manually or formulas vary between spreadsheets, it’s difficult to ensure outcomes are aligned across deals or over time.
Structuring how financial values are sourced, calculated and reused introduces consistency into the process. Lenders can trace results back to original documents, validate assumptions and apply policy evenly. That foundation supports stronger governance and makes decisions easier to defend long after approval.
How does automating administrative tasks free bankers to focus on building stronger client relationships?
Strong banking relationships are built on service, and service today looks very different than it did even a few years ago. Business borrowers expect timely responses, an experience that’s easy to navigate, and a clear understanding of how a lender arrived at a decision. When delays stretch on or explanations are vague, confidence erodes quickly.
When bankers aren’t consumed by file preparation and document rework, they have more capacity to engage borrowers directly. That means earlier conversations, clearer explanations around credit decisions and more thoughtful deal structuring. The way a lender evaluates risk and arrives at a decision is often their secret sauce and having the time and tools to explain that process helps reinforce trust. In a market where it’s easier than ever for borrowers to shop around, those moments of clarity and responsiveness matter.
How can AI give banks a competitive advantage in the increasingly tight commercial lending market?
Relationships have always been the competitive advantage in commercial lending, but they’re increasingly tested by expectations for speed and consistency. Borrowers don’t just compare rates, they compare experiences.
Institutions that can respond quickly, articulate their decisions clearly and move deals forward without unnecessary friction are better positioned to retain and grow relationships. When lenders are supported by clear processes that allow them to act decisively, they protect what differentiates them from competitors while meeting modern expectations. The combination of strong relationships reinforced by dependable execution is what separates leaders from followers in a crowded market.
What challenges do banks face when adopting AI, and how can they overcome them?
Hesitation is natural, especially in commercial lending. Credit teams have refined their processes over decades, and there’s a real concern that introducing AI could disrupt workflows, obscure decision-making or introduce new risk. Many institutions stay with familiar tools not because they work well, but because they’re understood. Over time, that comfort can quietly slow operations and limit growth.
Overcoming that hesitation starts with asking the right questions. Banks should be clear about what problem the AI is meant to solve and how success will be measured. They need transparency into how data is handled, whether models learn from their information and how outputs can be validated. In lending and risk contexts, explainability is essential. Lenders must be able to trace results back to source documents and understand how conclusions were reached.
Choosing the right partner is just as important as the technology itself. For banks, a strong partner understands banking operations, regulatory expectations and credit discipline. They work alongside lenders during adoption, embedding controls, reinforcing best practices and ensuring technology supports, rather than replaces, human judgment. When implemented thoughtfully, AI doesn’t disrupt lending, it actually strengthens it.
How do you see AI continuing to transform commercial lending in the next few years?
Over the next few years, commercial lending will increasingly shift from a specialized function to a core growth engine for many financial institutions. As banks look to expand beyond consumer lending, the ability to move confidently and efficiently in commercial credit will become a defining advantage.
AI will play a critical role in that shift by helping institutions move away from processes built around spreadsheets, manual data entry and disconnected systems. When lenders can rely on technology to read financial documents, apply underwriting calculations consistently and carry those values through origination, approval and review, they gain far more than efficiency. They gain confidence in their decisions, allowing institutions to scale commercial lending without scaling risk.
Faster decisions, stronger documentation and better visibility into portfolio health make it easier to grow responsibly, support business customers and stay competitive through changing economic conditions.
In 2026 and beyond, the institutions that treat AI as foundational lending infrastructure, not a bolt-on tool, will be the ones best positioned for long-term success.
What advice would you give to entrepreneurs, bankers, or technology leaders who are looking to drive innovation in their industries?
Get started. Research and planning are vitally important, but too many organizations, especially financial institutions, get stuck analyzing problems instead of addressing them. This means solutions get to market slowly or not at all.
Leaders should begin by assessing what capabilities they have internally and where outside expertise is needed. Most community financial institutions aren’t built to develop and maintain complex software, but they are experts in how they lend, how they manage risk and why their processes work for them. That expertise is incredibly valuable and shouldn’t be replaced.
