Construction Boots and AI

This is a cleaned-up transcript of a real conversation I had with my son Linus. He was not feeling well one morning, so he stayed home from school. I believe that spending time together during the day, asking questions and talking through real-world topics, is one of the best forms of learning. I recorded our conversation and used speech-to-text to capture it. What follows is the exact discussion, lightly edited for clarity and structure, but not embellished or fictionalized. Everything you read below is exactly what he asked and how I answered.

1. Linus asked, “Why do you have construction boots?”

I said, “Well, sometimes I have to go to a construction site for work.”

2. Then he asked, “But if you build AI software, why would you need to go to a construction site?”

I told him, “That is a really good question. It is the kind of question I might have asked when I was your age or even ten years ago. But now it feels completely natural. A huge portion of the global economy is built around physical construction. Not just building houses, but also roads, airports, bridges, offshore drilling platforms, manufacturing plants, refineries, and even mines. If you think of all those kinds of construction, you start to realize just how many people spend their days in those environments. It is a lot.

But here is the thing. Not everyone who should be wearing construction boots actually wears them. In many countries, if you walk onto a construction site, you will see workers wearing unsafe shoes. Many of them probably lose toes because of it. Safety gear like steel-toe boots matters more than people think. So yes, I build AI software. But the AI we create is not just some floating chatbot. It helps real industries solve real problems. For example, one of our clients might be using AI to assist maintenance inspectors who work in big factories or refineries. Those inspectors need to walk around giant machines, open panels, diagnose problems, sometimes in difficult or dangerous conditions.

Let me give you a more specific example. Let us say you are building a factory that produces toilet paper. When you buy a roll of toilet paper at the store, let us say it costs two dollars. You might think most of that cost is the pulp and paper used to make it. But actually, only a tiny part of the cost is the material. A much larger part is the capital cost, which is the money spent to build the factory. After that, there is the operational cost, which is keeping the factory running. But another cost that is just as significant as operations is maintenance.

And the people doing maintenance often need to be right there on the factory floor, checking the equipment, adjusting valves, climbing ladders, sometimes getting dirty or being around dangerous machinery. That is why they wear construction boots. And if I want to understand their job properly so I can build the best AI solution for them, I need to go where they work, see what they do, and sometimes wear the same boots they do.”

3. Linus looked down at my boots and asked, “How do they protect your toes?”

I said, “Good question. These boots have a steel cap inside them. You cannot really see it, but it is right in front where the toes are. If you were to step on my foot right now, you would not hurt me because the metal would stop your weight from crushing my toes. Some boots also protect the heel or are made to resist things like chemicals or extreme heat. Mine are fairly versatile. I could wear them at a refinery, at a construction site, or even a metallurgical plant. But if I had to work right next to molten metal, these would not be enough. You would need boots with extra layers that can protect you from splashes of extremely hot material. Every work environment has its own risks, and the right gear helps keep people safe.”

4. Then he asked, “So even though you work in AI, you sometimes need to follow people around and watch them do their jobs?”

I said, “Exactly. I am not in love with AI itself. AI is just a tool. What I care about is solving real problems. And to do that, I need to understand those problems deeply. That means sometimes stepping into other people’s worlds, like construction or manufacturing or maintenance, and seeing things firsthand. It is not enough to sit at a desk and imagine what they need. You have to go be with them.”

5. Then he asked, “What kind of problems are you solving?”

So I gave him another example. “Imagine you are a maintenance inspector. You go to a factory and see a huge machine that is vibrating weirdly. The machine was built fifteen years ago by a German company, and it is not something you are an expert on. You want to help, but you are not sure what to do. Maybe someone else will have to fly in from another city just to look at it.

But what if you had a head-mounted camera that is connected to an AI system? The AI could recognize the model of the machine. It could bring up the manual. It could guide you through a checklist, ask you to check the voltage, listen for certain sounds, open a panel and look at a gear. That would make you more efficient. You would not need to fly someone in. That is the kind of thing we build.”

6. Then Linus asked, “But if AI makes people more efficient, does that mean fewer jobs?”

I told him, “You are touching on a big debate. It feels like that should be true. If one person can do the work of three, then we need fewer people. But that is actually a fallacy. A fallacy means something that seems true but is not when you look more closely.

If we go back ten thousand years, almost everyone worked in agriculture. They farmed or raised animals. Then people invented new tools and techniques. Fewer people were needed on the farm. But those people did not just disappear. They started doing new kinds of work, like making pottery, crafting tools, or eventually working in factories.

It is the same thing today. AI is making us more efficient in farming, in manufacturing, and now even in services like accounting, legal research, or scheduling. That means we free up human talent to work on new problems.

And we have plenty of problems that need solving. We do not yet know how to harvest solar energy efficiently. Many people suffer from obesity. Kids are bored in school and not fulfilling their potential. Many people are depressed. Many still die young from cancer. These are real human problems. And solving them requires new knowledge. So AI does not eliminate work. It shifts us toward creating new knowledge and solving harder problems.”

7. Linus said, “But materials for building a business are more expensive now, right?”

I said, “That is true. Inflation means raw materials cost more. If you want to buy pulp and paper to make toilet paper, it is going to cost more than it did ten years ago. But remember, those costs go up for everyone, big companies and small ones. The big difference is that new businesses that use AI at their core can be much more efficient. They need fewer people, fewer layers of management, and less overhead.

For example, look at my company, bld.ai. We are a professional services company. We help clients solve problems and deliver knowledge. We compete with giants like Accenture and Infosys. Those companies are big, slow, and expensive. We are small, fast, and efficient. We deliver higher quality at lower prices, and we do it faster.”

8. Linus asked, “Then what is your biggest challenge?”

I told him, “Our biggest challenge is that people do not know who we are. We do not have a big brand. And in large companies, when someone chooses which vendor to work with, they often pick the brand they already know. If something goes wrong, they will not get blamed. But if they pick a small company and something goes badly, they might get in trouble. So we cannot just be slightly better than the competition. We have to be dramatically better. We have to be a lot faster, a lot cheaper, and much higher quality.

The good news is, when we get a chance to prove ourselves, we usually win. Over time, more people hear about us. My job used to be mostly about execution, delivering the work. Now, a lot of my time goes into marketing. I tell stories about our success, and I let our growing team handle the execution because they are excellent at it.”

9. Then Linus asked, “So you are not worried about doing the work anymore?”

I said, “Exactly. Our execution is rock solid. We have done more than five hundred projects. More than ninety seven percent of our clients pay on time. When someone does not pay, it is usually because they ran out of money, not because we did bad work. Especially with big companies, they always pay because we always deliver.”

10. Then he asked, “How do you market the company?”

I said, “I have to be patient and persistent. I tell stories like this one, conversations like the one we are having now. I turn them into blog posts. Then I share little pieces on platforms like LinkedIn, Twitter, Reddit, Instagram, and even TikTok. We also use email to reach people directly.”

11. Finally, Linus asked, “Is this why you sometimes use pictures? Is that for marketing?”

I said, “Yes, exactly. Words are the most important way to communicate ideas. But pictures come right after. A picture helps people see what we are talking about. For example, if I explain that someone is using AI to inspect a machine, I can also show a diagram of a person wearing a camera and looking at a big machine, with arrows pointing to different components. That makes it clearer. Or I can show a photo of me on site in my construction boots, talking to workers. That kind of image helps people trust that we know what we are doing.”

At that point, he nodded and smiled. I asked, “Is that the end of the interview?” He said, “Yeah.” I told him, “Thanks very much, Linus.”