Everyone keeps saying construction is behind on AI. I heard it straight from one of the guys I interviewed. He runs operations at a family-owned construction company, and when I asked him about AI in his industry, he said it plain: the construction world is pretty behind, it's just the nature of it.
And for a while I believed it too. It's the easy story. Old-school industry, guys in the field, not a lot of tech. Makes sense on paper.
But once I actually started talking to construction people instead of reading headlines about them, I stopped buying it. These guys aren't behind. They know exactly what they want AI to do. Nobody's built it for them yet.
Let me show you what I mean.
The safety director who already designed the tool
I talked to a corporate safety director at a crane and rigging company. I asked him the question I ask everyone: if AI could take one thing off your plate, no limits, what would it be?
He didn't give me a vague answer. He gave me three real ones.
First, a chatbot his field guys could text with safety questions. Stuff like "when should I use stop-work authority?" A lone worker out on a job with a real question and nobody around to ask.
Second, photo-based rigging checks. A rigger snaps a picture of a load and the AI tells him if it looks secure or if the configuration is wrong, before anyone gets hurt.
Third, auto-generated site-specific safety plans, so his team stops rebuilding the same document by hand for every new site.
This is not a guy who needs convincing that AI is useful. He's already mapped the whole solution. And here's the part that stuck with me. When I asked why he hadn't built it yet, he told me it's completely possible. The only thing stopping him is finding the right person to build it at the right price.
"It's possible. It's just getting the right person in for the right price and time. That's our struggle right now."— Corporate safety director, crane & rigging company
That's not "behind." That's a guy standing at a locked door with the key sitting just out of reach.
The operator who assumed it was impossible
The other conversation was with a director of operations at a construction company. His most repetitive work is processing invoices and cutting checks, every single Friday, over and over.
When I asked him his dream AI tool, he said accurate bidding. Being able to trust an AI to price a job. But he assumed it was near impossible, because every project is so unique.
And then he said the thing that I think most AI vendors completely miss. His guys in the field aren't tech-savvy, and they're slammed. Any tool that adds a few hours a week to learn is dead on arrival, no matter how good it is.
"A lot of our guys in the field are not going to be overly tech-savvy, and they're incredibly busy. Adding hours every week to learn a new tool is not easy for them."— Director of operations, family-owned construction company
So it's not that he's resistant. It's that the tool has to survive contact with a real job site, and most of them don't.
Not behind. Waiting.
Two totally different roles. Same story underneath.
The gap in construction isn't awareness. It isn't willingness. It's that the genuinely useful tools are specific, kind of unglamorous, and nobody's building them for this industry. And when someone finally does, it has to actually work for a busy crew that doesn't have time to mess around.
I'm going to keep talking to people in construction, because I don't think this is just two guys. I think there's a whole industry full of leaders who already know what they want and are quietly waiting for someone to build it.
— W.S., Boise, July 2026