The Headstart

I have spent nineteen years as an Autodesk Technical Specialist, which means I sit in front of a different civil engineering or infrastructure team almost every week. For most of that time, the conversation about new technology sounded roughly the same everywhere. This year it split into two conversations.

One kind of team is still treating AI as something to evaluate. A person gets an afternoon to try a chatbot. A committee is forming to write a policy. Nobody has been handed the time to actually build anything, so nothing changes about how the work gets done.

The other kind of team gave one person real time and real permission. Not a mandate, just room. And that person is now shipping small, unglamorous automations into the daily process: a script that used to be a checklist, a report that used to be a Friday afternoon, a review pass that used to wait for someone senior to be free. Nothing in that list is dramatic. Together it adds up to a team that moves through the same week differently than the team next to it.

I want to be careful here, because I know what this paragraph usually turns into. It is not a warning. I am not telling you to catch up before it is too late, that argument is made elsewhere on this site and I do not need to make it twice. I am telling you what I am watching happen, present tense, across a customer list I can actually name to myself if not to you.

And it turns out I am not imagining the shape of it. A PwC study this year found that 74% of the economic value AI is generating gets captured by 20% of companies, everyone else splitting what is left. BCG puts a finer point on the top end: about 5% of firms have reached what they call AI future-built, and those firms are seeing roughly five times the revenue increase and three times the cost reduction of the rest. Those numbers are not about my customers specifically. They did not need to be. I already had the anecdote. The studies just confirmed I was not the only one seeing it.

What strikes me is how early this shows up. It is not years of compounding advantage separating these two kinds of teams, it is one afternoon somebody said yes to, multiplied by fifty-two weeks. The headstart is still small enough to look optional. I can already point at who has it.

Be Better - Notice which side of the split you are on.