Most Small Businesses Are About to Fumble AI. I've Seen This Movie Four Times.

I've been building companies since the '80s. Started them, scaled them, sold a few, lost a few. Forty years of it. And if that much time teaches you anything, it's how to spot a pattern, including the one that's about to play out with AI.

Here's the pattern. Every time a real shift comes through (the PC, the web, mobile, social), most small-business owners do one of two things. Both are wrong.

The first group freezes. They wait to see how it shakes out. "Let's not chase the hype." That sounds like prudence. Usually it's just fear in a nicer suit. By the time it's obvious, the window's closed and the people who moved early own the ground.

The second group does the opposite, and honestly that one breaks my heart more. They buy the shiny thing, bolt it onto whatever they were already doing, and wait for magic. "We bought the AI. Now do something with it." Nothing happens, so they decide the whole thing was overblown.

Here's what forty years actually taught me, and it's boring on purpose: a business is a machine for turning judgment into cash. That was true before AI and it's true now. AI doesn't change the machine. It changes your leverage. Which means, and write this down: if your offer is wrong, AI just helps you make the wrong thing faster. As one of the sharpest operators I follow puts it, 10x of zero is still zero.

So the tool was never the point. The judgment is the point. The thing you have that the tool doesn't (knowing your customer, your offer, what "good" actually looks like) just became the scarce resource. A researcher named Ethan Mollick studies this for a living, and he said it better than I can:

"I suspect the people who thrive will be the ones who know what good looks like, and can explain it clearly enough that even an AI can deliver it."

Ethan Mollick, "Management as AI superpower"

Read that twice. The edge isn't the AI. The edge is you: the owner who's been in the weeds long enough to tell good from garbage and say so clearly enough to direct the machine. Forty years in the weeds turns out not to be a liability in the AI age. It's the whole advantage.

So what do you actually do? Not a twelve-month strategy. Today's AI is the worst you will ever use. It only gets better from here, so a long plan is just a long way to procrastinate. Instead:

• Pick the one thing that already makes you money. Not a new idea. The proven one.
• Put AI on that as leverage. Multiply a workflow that works; don't invent one that doesn't.
• Stay in the loop. You know what good looks like, so you're the quality check, not a bystander.
• Start small, this week, and watch the real number. Not logins. Not "we're using AI now." The actual outcome.

That's the whole game. The owners who win the next ten years won't have the best tools. Everyone will have the same tools. They'll be the ones who kept their judgment in the loop and pointed it at the right thing.

I'm betting my own next company on exactly that, and doing it in public, so you can watch me get it right and get it wrong in real time. Forty years in, starting over on purpose. Come along.