Tonight, while writing this, I built a complete business intelligence system.

Agenda management. Task tracking. Calendar integration. Email automation. A working system is ready to use tomorrow morning.

Took me maybe an hour while I was thinking through this story.

Thirty years ago, at Wavefront, that would have required a team of developers and months of work.

And that gap—between then and now—tells you everything you need to know about why AI doesn’t democratize entrepreneurship the way everyone thinks it does.

The Three Gates Nobody Talks About

I’ve been building companies for thirty years, and I’ve watched thousands of ideas die at predictable points. It’s not random. There’s a pattern.

Gate One: Having an idea worth pursuing.

Not just any idea—one that could actually become something people pay for. Most people never get past this. They have business ideas, the way other people have thoughts about organizing their garage. Interesting, but not real.

Gate Two: Turning that idea into an actual product or service.

This is where dreams meet details. Making the thing, figuring out costs, support, and distribution. All the unglamorous work that nobody posts about on LinkedIn.

Gate Three: Building a sustainable business that survives past year three.

Market fit. Sales. Operations. The long, grinding work of actually running a company.

Each gate kills most of the people who make it through the previous one.

The math is brutal, even if nobody tracks exact percentages. What I can tell you from three decades of watching this happen: very, very few ideas become sustainable businesses.

Yet people keep trying. Why?

Through Another Lens is a reader-supported podcast and publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

Fifteen Minutes That Changed Everything

I learned the answer thirty years ago in the most humbling quarter-hour of my professional life.

I was at Wavefront, working on Dynamation—the first interactive particle system generator. If you’ve seen smoke, fire, water, hair, or clothing in any movie from the nineties forward, you’ve seen particle systems at work.

We’d built something nobody had ever built before. Interactive control over complex simulations. Revolutionary technology.

Jim Hourihan had developed the original software at Santa Barbara Studios. We got it working, integrated it, and we were flying high. We could create effects that were impossible before.

Martin Plahn—our chief technical officer—called us into his office.

I walked in thinking we were about to get congratulated.

Instead, Martin looked at us and said, “Great technology. Really impressive work.”

Then he destroyed us.

“Where’s the documentation? Who’s writing the user manual? What’s the support model? How are customers going to learn this? What’s the training program? How does this integrate with the rest of our product line? What’s the pricing structure? Who’s the market? How big? What’s the sales strategy?”

Question after question after question. Detail after detail we hadn’t even considered.

We’d been in business for ten years. This wasn’t our first product. But we’d gotten so excited about making the technology work that we’d forgotten everything else required to make it a product.

Martin wasn’t angry—that’s what I remember most. He was genuinely surprised that we, experienced business people, were acting like kids with a science project.

“Just because you can get a thing to work,” he said, “doesn’t make it a product.”

The Lesson That Stuck

Dynamation became a huge product for us. Ten thousand dollars per license. Sustainable, profitable, successful.

But only because Martin forced us to think about all the stuff we didn’t want to think about.

The boring stuff. The detail stuff. The “how does this actually work as a business” stuff.

That moment taught me something I see playing out right now with AI: The technology is never the hard part. The hard part is everything else.

The One-Person Company

Fast forward to this year. Sam Altman reportedly said something that made me rethink everything: “We will live to see the first one-person billion-dollar company.”

If that’s true—and I think it might be—what does that mean for the three gates?

For the last two months, I’ve been building a virtual writer’s room. AI agents acting as writers, editors, researchers, fact-checkers, and coordinators. Each has specific roles, specific jobs, and specific quality standards.

This story started with an AI agent interviewing me. Then the writing team took over. Then the editors. Then the fact-checkers. All AI agents, all working together, all managed by me.

One person. An entire production team.

And tonight’s business intelligence system? That was just Tuesday. I needed something, I built it, I moved on.

What Everyone Gets Wrong

Here’s where everyone completely misses the point about AI and entrepreneurship.

People think AI eliminates the hard parts of building a business. Now, anyone can skip the boring details and jump straight to success.

But the hard parts didn’t change.

You still need to understand your market. You still need real customers willing to pay real money. You still need to solve actual problems, not theoretical ones.

You still need to think like Martin—asking all the questions nobody wants to ask about how this actually works as a business.

AI doesn’t eliminate any of that.

What AI eliminates is the excuse.

Students vs. Workers

My mom said something when I was a kid that I’ve never forgotten.

We had someone staying with us—I don’t remember all the details—but there was this big argument. My mom said this woman was always going to be a student because she never figured out what it was like to actually work.

She knew how to study. She was great at being a student. But she didn’t know how to work.

AI separates the workers from the students more clearly than anything I’ve ever seen.

The students are taking courses about AI entrepreneurship. Watching YouTube videos. Getting excited about frameworks. Studying possibility.

The workers are building things. Tonight. Right now. Solving real problems they actually have.

And here’s the uncomfortable part nobody wants to talk about: if you’re still studying instead of building, AI just made it really obvious.

The Gap Is Getting Wider

The workers are moving faster than ever. They’re building things that would have required teams. They’re testing ideas at the speed of thought. They’re iterating in hours instead of months.

Meanwhile, the students are still taking courses on how to use ChatGPT.

The gap between someone who studies entrepreneurship and someone who practices it? That gap just got massive.

AI doesn’t democratize entrepreneurship.

It reveals who was actually an entrepreneur and who was just interested in the idea of being one.

What This Means

If you’re reading this and you’re actually building things—not thinking about building things, but actually doing it—you know exactly what I’m talking about. You don’t need my encouragement. You’re already past the first gate.

But if you’re still waiting for the perfect moment, the perfect tool, the perfect idea?

The builders are already through gate three while you’re still studying gate one.

The three gates are still there. They’re still brutal. They’re still the same gates Martin taught me about thirty years ago.

And AI just made it really, really clear who’s willing to walk through them.

Thanks for reading Through Another Lens! This post is public, so feel free to share it. If you’d rather listen, you can on Apple or Spotify.

Get full access to Through Another Lens at marksylvester.substack.com/subscribe