Your Real Competitive Edge Isn’t Artificial—It’s Human Judgment

You've felt it too: That flicker of relief when a machine spits out in ten seconds what would've taken your team half a day … and the quiet worry that follows it.

Listen to your inner voice. Why?

Humans are so much smarter than the glorified search engines to which we apply the misnomer “artificial intelligence (AI).”

And science tells us we are only using a small portion of our brain’s capacity.

As for AI, it might be more appropriate to call it LI, or limited intelligence. Large Language Models (LLMs) basically do one thing; they predict the next word. And they invent information when their programming cannot solve the prompt with which they are being presented.

A colleague of mine told me a few weeks ago that while doing research on a complex project, Claude delivered some amazing statistics. When my colleague responded that the statistics didn’t fit, Claude responded that it made them up.

This isn’t to say that LLMs are not of value. How they think mirrors our own brains. It is an absolutely fascinating and brilliant design. They can do amazing things, if you train them and demand the truth.

They are supersonically quick at gathering and organizing information from anywhere in cyberspace—whether or not it’s copyrighted. And they can complete repetitive tasks ad infinitum, more accurately than humans and without fatigue.

Twelve hours to do a four-hour job

I also find them exhausting. Last year I conducted a research project for a client on a subject I have explored in prior years. The last time I did the research it took four hours to find reputable primary sources, cross reference the findings, and prepare the report.

Using an LLM took twelve hours.

It swamped me with data. It was fascinating and interesting, but it took hours upon hours to check the sources to make sure they existed and read the source documents to make sure that the LLM didn’t misunderstand or misinterpret the request or take the results out of context.

The report was robust with lots of attributions, but the substance was neither better nor more insightful than the work I had done previously. AI is getting smarter. Programmers are tweaking it at a feverish pace. And it’s learning from everything everyone is pumping into its gaping maw.

We’re betting big on machines and small on people

Companies are in a torrid race to build huge AI data centers and the blind plunge of investment to take advantage of its profitability is affecting the credit market for other companies in need of capital. Along the way we are losing sight of our most valuable asset—people and their potential.

Machines predict. People create.

I’m with Jacob Collier who said that LLMs only predict and are thus predictable, humans are not limited. Humans create new amazing things out of the mere vapors of imagination. And humans have hearts and passions and desires. We need to stop treating machines as if they are humans and humans as if they are machines.

To my point, on the one hand we are giving these LLM’s names and treating their ones and zeros calculations as personalities. On the other hand, as exemplified in a research study conducted by Julia Shin and Sandra J. Sucher, as reported in the Harvard Business Review:

Someone still has to catch the workshop

“Our interviews and research reveal that managers are drowning from being responsible for catching “workslop,” AI-generated content that looks professional, but lacks substance and fails to advance the actual task. They’re expected to validate AI outputs, identify errors, coach their teams in AI skills and core on-the-job principles, and uphold quality standards—all while facing unchanged or even increased delivery pressure and lacking formal support structures.”

Back to blue books

LLMs are also prompting us to be dumber, less curious, and lazy. Seeing interviews with college students admitting that they rely heavily on LLMs to do their research—and then trusting without verifying—should give us great concern.

A few months ago, while on a Zoom call with a faculty member from my alma mater, he told us how his exams are now in the classroom, reverting back to blue books and pencils and exam questions printed on paper. Students are not allowed technology in the classroom when they take exams.

Machines don’t give you attitude—people do

So why are we thoroughly enamored by complex search engines to get work done and solve our problems while devaluing people? Why are we so quick to cut staff and focus on machines? I think it’s pretty simple.

Companies exist to produce and sell things, and right now, that makes leaning on AI look like the more profitable move. People, meanwhile, come with emotions, egos, and personal styles that make them harder to manage. Machines don't give you attitude. They take correction without resistance. So you train them to do the job.

But if you're tempted to reach for AI instead of trusting your own judgment to hire a genuinely great person or manage a high-performing team, what do you stand to lose? More than this post can cover in one sitting.

Let’s be honest: People are worth the effort. I know—interviewing and managing performance can feel like risky business, but it doesn't have to be.

Hire for how they think, not what they know

If you haven’t leveraged it yet, formal axiology—the study of how people value things—closes the gap between what you need and what someone can actually deliver.

The Hartman Value Profile®—a tool built on axiology and something I’ve successfully used for decades with clients—doesn’t measure what someone knows. It measures how they judge, how they weigh a decision, how they see people and possibilities.

These are the human capacities no LLM can predict its way into. If you're serious about not losing your most valuable asset, start by hiring people whose thinking you can actually see.

AI will keep getting faster. It will never get better at judgment. That's still yours to build, and yours to hire for. Start there, and the rest takes care of itself.