The Biggest AI Mistakes Lenders and Brokers Are Making

Artificial intelligence is quickly becoming one of the biggest conversations in lending and fintech. Lenders and brokers are looking for ways to use AI to automate processes, improve efficiency, and scale their businesses.

But there’s a problem: many companies are approaching AI from one of two extremes.

On one side, businesses know they need AI but have no idea what they actually want it to accomplish. On the other, companies are trying to automate nearly everything overnight.

Both approaches can create problems.

So, what are some of the biggest AI mistakes lenders and brokers are making today?

Mistake #1: Implementing AI Without a Plan

One of the most common conversations we have at Centrex Software starts with a company asking about AI.

“Do you have AI?”

Our next question is usually: What are you trying to accomplish with it?

And sometimes, there isn’t an answer.

AI has become such a major buzzword that businesses know they should be paying attention to it, but that doesn’t necessarily mean they have identified a specific problem they want AI to solve.

That’s where companies can run into trouble.

Before implementing AI, lenders and brokers should first think about what they actually want the technology to accomplish.

Are you trying to automate certain workflows? Do you want AI to help your team analyze information? Are you looking to improve communication? Do you want to introduce agentic AI into specific processes? Or do you simply want to start with one use case and expand from there?

The technology should follow the objective—not the other way around.

When you approach a software provider with a clear understanding of the problem you’re trying to solve, it becomes much easier to determine where AI can provide meaningful value.

Mistake #2: Trying to Automate Everything With AI

At the opposite end of the spectrum are lenders and brokers who have gone all-in on AI.

They’re automating emails. They’re automating underwriting. They’re automating portfolio management. They’re trying to remove human involvement from as many processes as possible.

The assumption is that AI can take over nearly everything and create a completely automated business with very few employees.

But AI isn’t necessarily ready to replace every process—or every person.

When too much is automated too quickly, businesses can lose something incredibly important: the human element.

Lending is still a relationship-driven industry. Borrowers, brokers, lenders, partners, and investors often want to communicate with actual people, especially when situations become complicated or require judgment.

AI-generated communication can also feel artificial when systems haven’t been properly trained, implemented, or monitored. Instead of making a business more efficient, excessive automation can create new problems that ultimately require teams to backtrack.

AI Should Be a Tool, Not the Entire Business

The goal shouldn’t necessarily be to determine how many employees AI can replace.

A better question is: How can AI help your existing team operate more effectively?

AI can be an incredibly powerful tool for lenders and brokers when it’s applied strategically. It can assist with repetitive processes, help teams work through information more efficiently, support automated workflows, and potentially reduce the amount of manual work required throughout an organization.

But that doesn’t mean every process needs AI. Some areas of your lending or broker business may benefit tremendously from automation, while others may still require human judgment, expertise, communication, and relationship building.

Finding that balance is critical.

Take a Crawl, Walk, Run Approach to AI

For lenders and brokers interested in implementing AI, a crawl, walk, run approach can make much more sense than trying to transform the entire organization overnight.

Start with a specific problem.

Implement AI in a controlled area where there’s a clear objective. See how the technology performs. Determine how your employees and customers respond. Identify what’s working and what needs improvement.

Then expand.

As your organization becomes more comfortable with AI, you can introduce it into additional workflows and processes where it makes sense.

This allows your AI strategy to evolve alongside your business rather than forcing your entire operation to adapt to new technology all at once.

Finding the Right Balance With AI in Lending

The biggest AI mistakes lenders and brokers are making can ultimately be found at two opposite ends of the spectrum.

One is implementing AI without knowing why. The other is implementing too much AI, too quickly.

The better approach falls somewhere in the middle.

Develop a plan. Identify the problems you want AI to solve. Start with specific use cases. Keep people involved where human judgment and relationships matter. Then gradually expand your use of AI as you learn what works for your organization.

AI doesn’t have to replace the way your lending business operates.

Used strategically, it can become a tool that helps your team operate smarter, faster, and more efficiently.