The Relevance Gap

You've survived technology changes before. The web. Mobile. The cloud. Every time, you had time. If a competitor moved first, you could watch, learn, and catch up a year later. No real damage. Being a fast follower was smart.

That's over.

AI doesn't work like the shifts before it. The company that moves first doesn't just get a head start. It gets further ahead every month, faster than you can catch up. I call this the relevance gap: the distance between businesses already using AI and those still working the old way. It widens faster every month you wait.

This article is about why that happens, how fast, and what waiting actually costs you. Start here.

AI isn't a new tool. It's a reset of who's competitive.

Every ten to fifteen years, technology doesn't just improve. It resets who wins. Mainframes lost to the PC. The PC lost to the web. The web lost to the smartphone. Each reset minted new winners and quietly retired the incumbents who thought their lead was permanent.

These resets are brutal. When computing moved into everyone's pocket, Microsoft fell from over 90% of computing to under a quarter in a decade. They didn't make a mistake. The ground moved, and they were standing where it used to be.

AI is the next reset, and it's early. Most industries are still at the starting line: legal, accounting, healthcare, logistics, real estate. The intelligence itself is becoming a commodity, something you buy like electricity. Sam Altman calls it “intelligence as a utility, on a meter.” When raw capability is cheap and available to everyone, owning AI is worth nothing. Using it inside your business before your competitor does is worth everything.

That's what most owners miss. They treat AI as a one-time software purchase. It isn't. It's a shift in who gets to compete at all.

Why you can't catch up later

Here's the mechanism. It's the whole reason “wait and see” stops working.

AI isn't a one-time bump. A new website or CRM gives you one step up, then you plateau. AI compounds. A business running on AI ships faster, so it learns faster, so it improves faster. Each loop feeds the next. The competitor who started six months ago isn't six months ahead of you. They're on a steeper curve, and it gets steeper the longer they run on it.

So the gap doesn't just grow. It accelerates. In past shifts, the distance between leader and follower stayed roughly flat, and you could always sprint and close it. Not this time. This gap widens every month, and every month it's harder to close than the last.

This isn't theory. It's in the earnings reports.

The clearest signal isn't in predictions. It's in what profitable, growing companies are doing right now.

Jack Dorsey cut nearly half of Block's staff, close to 4,000 people, while the company was making almost $3 billion in gross profit in a single quarter. He didn't do it to survive. He did it because AI made it possible to do more with fewer people. The stock rose 23%. His words:

“Within the next year, I believe the majority of companies will reach the same conclusion and make similar structural changes.”

He's not an outlier. And these aren't failing companies:

• Dell has cut 36,000 jobs since 2023, including 11,000 in early 2026. Stock up 24% on the year. AI server revenue up 342%.
• Cloudflare cut 1,100 people in a record quarter: $639.8M in revenue, 34% growth.
• ClickUp cut 22% of its staff and called the business “the strongest it's ever been.”

The ClickUp CEO put it better than I can:

“I only see two options: wait for this to play out gradually in the market, or be honest about what I'm seeing and act proactively. Waiting for that to happen naturally means falling behind now.”

The pattern is simple. Fewer people, more AI, higher output, and the market rewards it. That's the relevance gap in practice. The companies that moved aren't just surviving. They're pulling away from everyone who didn't.

This goes beyond big-tech layoffs. A 2026 Cognizant study found AI can now affect 93% of US jobs. Mustafa Suleyman, who runs Microsoft AI, said in early 2026 that most professional tasks will be automated within 12 to 18 months. Argue with the timeline if you want. You can't argue that the people building this and the people running profitable companies are all moving the same direction at once.

How fast this is moving

Most owners accept that AI is changing things. They underestimate the speed every time. This capability doesn't improve in a line. It compounds, just like the gap. Watch it move:

• 2022. AI couldn't do basic arithmetic. It would tell you, confidently, that 7 × 8 = 54.
• 2023. It passed the bar exam. Still useless for real work, but the direction was clear.
• 2024. It wrote working software and explained graduate-level science. It started taking entry-level tasks.
• Late 2025. The best engineers alive handed it most of their work. Benchmarks measured it finishing five-hour expert tasks on its own.
• Now. It builds entire applications end to end, tests its own work, and improves with no person in the loop. CEOs of leading companies say it does “much of the work.”

None of this is a lab demo. It's running in real companies today. And the people building it say it speeds up from here.

The math nobody wants to do

Because the gap compounds, waiting costs far more than it feels like it should. Say a business that starts today has an advantage of “1.” Here's what happens to the one that waits while the first keeps compounding:

These aren't made-up numbers. When a competitor ships in days instead of months, answers customers in real time, and improves faster than you can finish a planning meeting, the compounding is real. You measure it in market share, margins, and headcount.

What it looks like for one real business

Make it concrete. Two accounting firms, same city. One starts using AI in early 2026 to automate compliance checks, draft reports, and read client data. The other waits to see how it plays out.

Two quarters later, the first firm handles 40% more clients with the same team. A year later, it launches an AI-built advisory service the second firm hasn't even imagined. Notice what changed. The gap stopped being about speed and became about capability. The second firm isn't slightly behind. It's in a different league. And it never saw the moment it lost, because the race never looked like a race.

“Wait and see” is the one strategy that fails

Here's the uncomfortable part, and it's why I'm writing this.

“Wait and see” works against slow change. Let others go first, watch what works, copy it. It protected businesses for decades. Against this kind of change it's the one approach guaranteed to fail. By the time you can see the results clearly enough to copy them, the gap is already too wide to close. The strategy that kept you safe through every past shift is the exact thing that sinks you in this one.

There's real opportunity in that. Most businesses, especially small and mid-sized ones, haven't started. The window is narrow, but it's open. Closing the gap doesn't mean hiring an AI department or spending millions. It means finding where AI actually fits your business and starting there. Not buying more tools. Everyone has the same tools now. Find where they create real impact in how you already work, and act before your competitors do.

That's the part I find genuinely hopeful. The leaders aren't ahead because they bought something nobody else can. They're ahead because they started. The intelligence is the same for everyone. The only thing separating the businesses that stay competitive from the ones that quietly fall out of relevance in the market is whether they move while the window is open.

That's the whole game right now.

This is part of a series on what AI is doing to how we build and run businesses. For the view from inside product teams, see The New Product World on how AI is collapsing traditional roles, and The Four Levels of Agentic Product Teams on how deeply AI is reshaping how teams work.