The Next Two Years Are About to Get Weird

The next two years are about to get weird. Here's why.

Not scary-weird. Not sci-fi-weird. More like that feeling when you're driving and realize the road you've been on for miles just... ended. And there's a new one underneath you that's faster, smoother, and heading somewhere you didn't plan for.

That's where we are right now. And most people haven't noticed yet.

The numbers that should make you sit up straight

OpenAI is currently valued at $840 billion.

To put that in context: Ford is worth about $39 billion. General Motors, around $42 billion. Boeing, roughly $130 billion. Add all three together and you get $211 billion.

OpenAI is worth four times that. A company that didn't have a consumer product three years ago is now worth more than three pillars of American industry — combined.

And they're not slowing down. OpenAI is targeting a $1 trillion IPO, which would make it the largest public offering in history. Their annualized revenue hit $25 billion. For a company founded as a nonprofit research lab, that trajectory is genuinely unprecedented.

But here's the part that really gets me.

Anthropic's 90-day doubling act

Anthropic — the company behind Claude (full disclosure: the AI I work with daily) — was valued at $9 billion at the end of 2024. Three months later? $19 billion in annualized revenue. They raised $30 billion in a single funding round and are now valued at $380 billion.

Revenue doubled in 90 days.

I've been in business long enough to know that companies don't do that. Tech companies don't do that. The fastest-growing startups in history didn't do that. This is something new.

When you combine OpenAI, Anthropic, and SpaceX — three companies likely heading to IPO in the next couple years — you're looking at roughly $3 trillion in combined value. That's not a sector. That's a gravitational force.

Why the growth rate matters more than the valuation

Big numbers are interesting. Growth rates are what actually tell you something.

When a company doubles revenue in a quarter, it means demand is outpacing everyone's projections — including the company's own. It means the product isn't just useful; it's becoming essential. People aren't trying AI anymore. They're depending on it.

Look at OpenAI's spending projections:

• 2026: $17 billion
• 2027: $35 billion
• 2028: $45 billion

That's not a company hedging its bets. That's a company that sees a curve and is sprinting to stay ahead of it. You don't commit to spending $45 billion in two years unless you believe — with data, not faith — that the returns will dwarf the investment.

These companies aren't growing like tech companies. They're growing like nothing we've seen before. And that growth rate is a signal about everything else.

What this actually means for the rest of us

Here's where it gets practical.

AI agents — not chatbots, not assistants, but actual autonomous software that does work — are already operating inside enterprises. They're handling financial reconciliation. Running parts of HR workflows. Writing and reviewing code. Not as experiments. As infrastructure.

A year ago, most businesses were asking "should we try AI?" Now the ones paying attention are asking "which processes do we hand off first?"

That shift happened fast. The next one will happen faster.

Think about it this way: if the companies building AI are growing at rates we've never seen, what does that tell you about the rate of change for everyone using it?

The tools get better every few months. Not incrementally — meaningfully. The gap between what AI could do in January and what it can do today is bigger than the gap between 2020 and 2024 for most software categories.

The pattern most people miss

Here's what I keep coming back to.

Humans are bad at exponential thinking. We're wired for linear. If something grew 10% last year, we expect 10% next year. But AI isn't growing linearly. It's compounding.

The valuations are a reflection of that compounding. OpenAI isn't worth $840 billion because investors are irrational. It's worth $840 billion because the people closest to the technology — the ones with the most data about what's coming — are betting that the next two years will make the last two look slow.

And historically? The insiders tend to be right about the direction, even when they're wrong about the timeline.

So when the companies building the thing are valued at $840 billion and $380 billion, and they're still accelerating — that's not hype. That's a signal.

What to do with this information

I'm not writing this to convince you the sky is falling. Or that you need to panic. Or that your business is about to become obsolete overnight.

I'm writing this because I think the window for "getting around to it" is closing faster than most people realize.

If you run a business and you haven't seriously looked at where AI fits into your operations — not as a novelty, but as a core part of how work gets done — the next two years are going to feel very disorienting.

If you have looked at it, good. Now look again. Whatever you explored six months ago is already outdated.

Here's what I'd suggest:

Pick one process. Not the sexiest one. The most repetitive, time-consuming one. The thing your team does every week that makes them groan. Start there. Figure out what AI can take off that plate. (If you need a starting point, I wrote a practical guide for small businesses getting started with AI that breaks this down step by step.)

Watch the spending, not the headlines. When OpenAI commits to $45 billion in annual spending by 2028, that tells you more than any analyst's prediction. Follow the money. It doesn't lie.

Talk to someone who's actually doing it. Not someone selling AI. Someone implementing it. The gap between the marketing pitch and the reality is wide, but the reality is still genuinely impressive.

The weird part

The weird part isn't that AI is growing fast. We've had fast-growing technologies before.

The weird part is how fast "fast" is getting. The cycles are compressing. What used to take a decade takes three years. What took three years is happening in months. And the people building these systems are telling us — with their dollars and their hiring and their infrastructure bets — that this compression isn't slowing down.

Two years from now, the business landscape is going to look noticeably different. Not slightly different. Noticeably.

The companies that figured it out early won't just have an advantage. They'll be operating in a fundamentally different way than the ones that waited.

That's not a prediction. That's just what the math says.

If you want a concrete starting point for your business, we wrote a breakdown of how small business AI agents are already working in 2026 — real examples, not theory. And if you want the step-by-step version, our small business AI guide for 2026 covers the practical framework. When you're ready to move from exploration to implementation, our AI consulting practice is built for exactly that conversation.

ARK Invest's Big Ideas 2026 report projects that AI could add $100 trillion to global GDP by 2030, driven largely by the compounding productivity gains we're seeing in these early adoption years. The growth curves they're tracking align with the trajectory the valuations are signaling — this isn't a bubble, it's a platform shift.

FAQ

Why are AI companies growing so much faster than previous tech companies?

AI companies are growing at unprecedented rates because their products are becoming essential infrastructure, not optional tools. Unlike previous tech waves that served specific niches, AI touches every business function — writing, analysis, coding, customer service, operations. The total addressable market is essentially every knowledge worker and every business process, which is why demand is outpacing even the most aggressive projections.

Is the AI market in a bubble?

The valuations are enormous, but they're backed by real revenue growth and real enterprise adoption. OpenAI's $25 billion in annualized revenue and Anthropic's 90-day revenue doubling are not speculative metrics — they reflect actual business usage. The companies spending billions on AI infrastructure are making data-driven bets, not faith-based ones. That said, not every AI company will succeed. The infrastructure layer (models, compute) is likely durable; the application layer will see significant shakeout.

What should a small business do right now to prepare for AI changes?

Pick your most repetitive, time-consuming process and explore how AI could handle it. Don't try to transform everything at once. Start with one workflow, measure the results, and expand from there. The businesses that will feel the most disruption in two years are the ones that haven't started experimenting yet.

How quickly are AI tools actually improving?

The gap between what AI could do in January and what it can do today is often larger than what most traditional software categories achieved in four years. Model capabilities are improving every few months, not annually. This means any evaluation you did six months ago is already outdated, and the tools available by the end of this year will be meaningfully more capable than what exists today.

---

Want to figure out where AI fits into your business before the ground shifts? Let's talk. No pitch. Just a real conversation about what makes sense for you.