Artificial Intelligence vs. Automation Intelligence
By David J. Saxton, President, The Computer Manager
What’s another word for “artificial”? Take a moment. What was the first word that came to mind?
If you said “fake,” you’re not alone. I’ve asked this question to nearly 100 people, and every single one gave the same answer. Coincidence? I don’t think so.
Now, don’t get me wrong—Artificial Intelligence is one of the most remarkable tools humankind has ever created. The speed at which it’s evolving is nothing short of extraordinary. But there’s a problem I keep seeing: too many people are treating AI as a solution.
That’s a mistake.
My team and I use AI every day. But from the beginning, we’ve been intentional about one thing—we view AI as a tool, not a solution. We don’t allow AI to make decisions for us. Instead, we ask it to propose solutions. Then we step in.
We analyze. We question. We refine.
If AI gets it right, great. If it doesn’t, we provide more direction, more context, and try again. Only when the output meets our standards does a human make the final call. That human layer—discernment, judgment, accountability—is essential. And it’s where the real value is created.
Let me give you a simple analogy.
Before the chainsaw, people still cut down trees. It just took a lot longer. Then the chainsaw arrived—and everything changed. A task that once took 20 hours could be done in minutes. It was a massive leap in efficiency.
But in the wrong hands, a chainsaw can cause serious damage.
AI is no different.
It’s an incredibly powerful tool. But it’s still just a tool. And like any tool, its value—and its risk—depends entirely on how it’s used.
The reality is this: AI still lacks something critical— discernment. It can process enormous amounts of information and generate impressive outputs, but it doesn’t truly understand nuance, context, or judgment the way humans do. At least not yet.
That’s why, in our shop, we treat everything AI produces as a suggestion—not a conclusion. We apply human judgment to determine the right path forward. That approach has made a measurable difference for us.
Let me share an example.
A few years ago, I began a deep dive into every role within my collection agency. The goal was straightforward: identify tasks that humans were performing and determine which of those could be automated using machine learning.
At the time, we had 80 full-time employees. Over the course of the project, we developed automation that replaced a significant amount of manual labor. In total, we implemented hundreds of process improvements.
Here’s one example.
We work with a credit card client that sends us monthly account statements before we begin legal action. These statements are delivered as individual PDF files—usually 12 to 36 per account.
Previously, an employee would manually sort and combine those files into a single document. It was tedious, time-consuming work.
When I discovered this, we built automation to handle it.
Once the programming was complete, we had a backlog of about 4,500 documents. We estimated it would take roughly 100 employee hours to process manually. Our automation completed the task in nine minutes.
Nine minutes.
We’ve been building this kind of intelligent automation— inside and outside of our collection platform, Debt$Net— for the past three years. And the results have been significant.
Today, we operate with 50 employees. Over the last 12 months, we’ve posted record revenue and record profit margins—without adding a major new client.
So what do we call this approach?
Automation Intelligence.
We use Artificial Intelligence as a tool to build Automation Intelligence. Both involve machine learning. The key difference is that Automation Intelligence keeps a human in the decision-making role—and that’s the critical piece.
In our industry, “close” isn’t good enough. The risk is too high. Until machines can truly replicate human discernment, we’ll continue to rely on people to make the final decisions.
Our focus remains simple: use both AI (artificial intelligence) and AI (automation intelligence) to create measurable outcomes—reduced costs, increased efficiency, and higher revenue.
If a vendor is talking to you about AI, ask them a simple question: What has it actually done?
If they can’t point to real-world results—lower headcount, reduced expenses, increased revenue—then it is probably just hype. We’ve seen this before. Think back to the dot-com boom and bust.
We’re not chasing the AI frenzy.
We’re being deliberate. We’re using AI to improve efficiency, reduce costs, and grow revenue—responsibly and strategically.
Both types of AI.
And that’s exactly how we’ll continue building the future of DebtNet$Millennium.
So stay tuned. We’re just getting started.