You Cannot Lead Them Where You Have Not Gone

Thirty years of leading people is the half you have. It is not the half that will carry them across.

You lead people. You command technology. Those are two different jobs, and the second one is not optional any more.

It does not matter what the people are called. A crew of twelve in the field. A department. A practice with four partners and a nurse. A firm, a shop, a shift, two associates and a paralegal. If other men's work carries your name on it, this is you.

And if you are running alone, starting something new or starting over, this is you twice over. The twelve jobs your old outfit spread across twelve people all land on you now, and the only way you cover that ground is by commanding tools that work while you sleep. For you the two halves collapse into one. The technology is the crew.

Said plainly before the evidence: if you are good with people and have not taken command of the tools, you cannot lead them through this. Not because your leadership is weak. Because you cannot lead anyone to a place you have not been yourself. They can feel it when you are guessing, and they will not follow a guess.

This matters this month. Gartner, the technology research firm, forecasts that forty percent of enterprise applications will have a task-specific agent built in by the end of this year, up from five percent late last year. That is a projection. This is not: IBM reports running roughly four thousand digital workers alongside its people right now. The agents have arrived at work.

The tools are not where it fails. Microsoft ran the numbers on twenty thousand workers this year and found that about two thirds of what an organization actually gains from this technology comes from culture, from managers, from how the work is arranged around it. Not from any individual's skill with the software. The agent can run perfectly and the whole thing can still come apart, because nobody redesigned the roles or prepared the people around it.

Unled people do not stay neutral. In one enterprise survey this year, twenty-nine percent of employees admitted they had actively worked against their own company's AI strategy. You have run people. Read it that way. That is not laziness and it is not stupidity. That is what a man does when something lands on him that nobody explained, nobody asked him about, and nobody led him through. He protects himself. You probably would have done the same.

And there is a bill only the man in charge gets. The work still goes out under your name. The associate drafts it with a tool you have never opened and you sign it. The resident brings you a read the software flagged and you approve it. Your estimator prices a job with a model nobody has audited and you shake on the number. You can supervise output you cannot evaluate for exactly as long as nothing goes wrong. Then it is your signature on it, and your license, and your reputation, and the honest answer to how did this happen is that you never looked.

Leading and commanding are not the same skill. Leading runs on things a machine does not have. Trust, persuasion, morale, your read on what a man is capable of before he knows it himself. None of that applies to an agent, because an agent cannot be inspired and will never be loyal. What the machine needs instead is clear intent, a defined outcome, a boundary on what it may touch, and your judgment checking its work. Half of that you already learned running people. The other half you have to go get.

Here is the comfortable read, and it is the one most men our age are taking.

I am a people guy. I have always been a people guy. The technology is for the young ones to figure out, and when they have it sorted, I will do what I have always done, which is lead the men who use it.

It sounds like wisdom. It sounds like knowing your lane. The partner leaves it to the associates. The physician leaves it to the residents and the vendor rep. The owner leaves it to whoever in the office seems to understand computers. Three different collars, one identical decision.

And it fails for one reason that has nothing to do with your ability with people: your people will take the tools seriously in exactly the proportion that you do. If the man at the top has not put his hands on it, everyone below him learns it is optional, or that it is somebody else's problem. You did not tell them that. You showed them.

That is the trap. Delegating the command and keeping the leadership feels responsible, and it strands everyone. Your people are led beautifully to the edge of a thing their leader has never crossed, and there they stop.

I led people for thirty years and I was good at it. That is not a boast. It is the setup for the part that matters.

When the technology started reshaping my industry, I did what a people guy does. I kept leading. I held the culture together, I kept the men steady, I made the calls I had always made. What I did not do was take command of the technology myself, and because I had not, I could not lead anyone else through it.

It was both sides of the house. The office never got led around and through what was coming, and neither did the field. Two crews, two different relationships to the tools, and I brought neither of them across. My company became an early statistic.

That falls on me. It is also not lost on me.

Element 13 of The Sovereign's Code, from the book I am finishing, is FORCE MULTIPLICATION:

Achieve exponential impact beyond your personal capacity by building systems, activating people, and commanding technology.

Activating people and commanding technology. Both of them, in one sentence, joined by an and. I had spent my whole career on the first one and treated the second as somebody else's department, and the Element says plainly that it takes both to multiply anything.

A man running on one of the two is not multiplying. He is just working hard, with a good crew, in the wrong decade.

Go first, then bring one man with you.

Pick one real task in your own work and hand it to an agent this week. Not a demo, not a test question. Something that actually has to get done, where you will know whether the output is any good because you have done that task a hundred times. Give it clear intent, define what finished looks like, then check the work the way you would check a new hire's.

You are not trying to become technical. You are getting your own hands on the thing so that when you open your mouth about it, you are speaking from the far side of it instead of guessing.

Then do the leadership half. Pick one person whose work carries your name, an associate, a nurse, a foreman, a junior partner, and walk him through what you just did. Show him the part that worked and the part that did not. Tell him what you are still unsure of. That is not weakness in front of someone you outrank, it is the only thing that makes a man believe he is allowed to learn something in front of you. One person this week. That is how people get brought across, one at a time, by somebody who has already been.

If you are running alone, the second half is different but it is not lighter. Write down the one function you have been doing badly because there is nobody else to do it, the follow-up, the books, the proposals that never go out. That is your first hire. Go command it.

The research coming out this year reads, to me, like a report on my own company written a few years too late. Unled people working against the change. Culture and managers deciding the outcome while everyone argued about software. A leader who was good with people and had not gone first.

You still have the harder half. Thirty years of knowing how to move people is not something the machines are coming for, and most of the outfits deploying these tools do not have it and cannot buy it.

Go get the other half. Then take them across.