AI Is Everywhere. The Work Is Mostly the Same.

The technology became ambient. The organization stayed the same shape.

That gap, between how much the technology has spread and how little the structure around it has shifted, is what makes this moment slightly strange.

Most adoptions don't look like this. When email landed in companies in the 1990s, the org chart eventually adapted. When mobile devices became standard, sales organizations reshaped themselves around them. When cloud infrastructure matured, IT and finance restructured how they thought about cost and ownership. Tools usually precede structural change, sometimes by years. But change comes.

The peculiarity of this moment is the speed of the technology relative to the inertia of the structure. AI capabilities have advanced inside organizations at a pace none of those earlier waves moved at. The structures around them haven't kept up. Or haven't yet.

This isn't a complaint about adoption

It's worth being precise about what this observation is and isn't.

It isn't a complaint that AI hasn't transformed enough yet. The early phase of any technology wave looks like this. It isn't an argument that organizations are doing it wrong. Most of what's happening right now is exactly what early adoption is supposed to look like. Tools spread. People learn what they're for. Productivity improves at the individual level. The broader structure waits to see what's worth changing.

It also isn't a prediction. Plenty of confident voices are willing to tell you exactly how the org chart of the future will look. We aren't going to be one of those voices. Predictions about how this resolves tend to age poorly.

What we'd say instead is something narrower. Two true things are sitting next to each other right now, and the gap between them is going to get harder to ignore.

A question that's starting to surface

In a lot of conversations with leaders we've talked to over the last several months, a similar question keeps coming up in slightly different forms.

It usually sounds something like, "We're getting real productivity gains at the individual level, but I can't really point to where that shows up at the company level." Or, "Everyone's using AI, but the company doesn't feel different." Or, "We've spent two years getting better at the tools. We haven't really spent time figuring out what they change about how we work."

The honest answer to all three of those is the same. The technology has shifted faster than the operating model. Output has gotten cheaper. The work itself has mostly not moved.

That's not a bad place to be. It is a place that's getting harder to stay in.

Where this is going

Most organizations using AI today are still asking how to get more of it into their existing operations. That's a reasonable question for the early phase. It also has a ceiling, and the data is starting to make the ceiling visible.

A different question is sitting underneath, and most organizations haven't started asking it yet. Not what AI changes about the existing work, but what the existing work would look like if it had been designed for AI to be in it.

That's not a question we'll try to answer here. It's worth naming, and it's the question we're going to spend most of this year coming back to.

For now, the observation is enough. AI is everywhere. The work is mostly the same. Both of those things are true. The gap between them is going to be the story of the next two years.