The $189 Billion AI Boom Isn't Bullish. It's Herd Behavior.

Global venture funding hit $189 billion in February, the largest startup funding month on record. Most people saw that number and read it as strength. Proof the AI market is broad, healthy, and getting deeper by the week.

That is not what the number says. The real number is 83%. That's how much of February's record venture funding went to three companies: OpenAI, Anthropic, and Waymo. That isn't a boom spreading across a frontier. It's a crowd stampeding into the few names nobody has to defend.

Søren Kierkegaard, the 19th-century Danish philosopher who spent much of his work attacking conformity and crowd-thinking, had a line for this: the crowd is untruth. Not because every person inside it is wrong. Because once enough people move together, judgment gets replaced by borrowed certainty. People stop asking what's true and start asking what is safest to believe.

That's what this market feels like.

The headlines say AI is exploding. The deeper truth is narrower and more revealing: capital is not spreading across a frontier. It's piling into the few names nobody will get fired for backing.

And founders are in danger of learning the wrong lesson from it.

The crowd never buys truth first

The comforting story about venture capital is that it exists to discover the future. A bunch of sharp people scanning the edge, noticing what others missed, then funding conviction before consensus arrives.

Sometimes that's true.

But late in a cycle, the machine changes character. It stops being discovery and starts being defense. Money no longer asks, "What is true that few people see?" It asks, "Where can I hide inside the thing everyone already agrees matters?"

That is not the same behavior.

A market where 83% of capital goes to three companies is not merely excited. It's compressing fear into a small set of names and calling the compression wisdom.

That's the part most founders miss.

The crowd is not drawn to truth. It's drawn to relief.

Relief from uncertainty. Relief from having to explain an unconventional bet. Relief from the pain of standing alone if the thesis breaks. If the whole market is running toward the same three companies, then individual investors can outsource responsibility to consensus. If it works, they were right. If it fails, at least they failed with everyone else.

Kierkegaard would have recognized the move instantly. The crowd doesn't remove risk. It removes the feeling of personal authorship over risk.

That's why it feels so good.

And why it's so dangerous.

Founders copy the capital long before they notice they have

Most founders think capital concentration is mainly an investor story.

It isn't.

It becomes a founder psychology story the second you start adapting your ambition to match the crowd's map.

When every headline, every podcast, every funding roundup, and every dinner conversation revolves around the same few companies, you start absorbing a quiet instruction: build what resembles what the market already blesses.

Not consciously.

That's what makes it lethal.

You don't say, "I'm going to imitate the crowd now." You just begin trimming your imagination to fit the current center of gravity. You over-index on what sounds frontier-lab-adjacent. You start talking in the language investors reward. You pitch proximity to the winners instead of clarity about the actual wound you solve.

Soon your company has a story the market understands and a soul you barely recognize.

This is how founders become derivative while feeling strategic.

The danger in an AI gold rush is not only that money floods the obvious names. It's that everyone outside those names starts reorganizing themselves around the wrong source of truth.

You begin to think scale is the moat. That compute is the moat. That being close to the center is the moat.

Sometimes those things matter.

But most real companies are not built by winning the same war as OpenAI. They're built by noticing what the crowd's fixation makes invisible.

Every stampede creates its own blind spot

Here's the strange gift inside a market like this.

When the crowd gets obsessed with one layer of the stack, it neglects the others.

If everyone is staring at model labs, they miss workflow ownership. If everyone is staring at compute, they miss trust. If everyone is staring at foundation models, they miss distribution. If everyone is staring at valuation, they miss behavior.

That's where the edge usually is.

Not in pretending the center doesn't matter. It does.

But in understanding that the center distorts attention. And distorted attention leaves entire categories of opportunity underbuilt.

The founders who win in an era like this are rarely the ones who panic hardest at the size of the incumbents. They're the ones who understand that giant concentrations of capital create giant concentrations of blindness.

This is true in every cycle.

The loudest money teaches the market what to notice. It also teaches the market what to ignore.

Kierkegaard's deeper point wasn't just that the crowd gets things wrong. It was that the individual has a duty not to surrender his own interior judgment to the comfort of collective agreement.

That is not just a moral point.

For founders, it's a strategic one.

Because the second you let the crowd tell you where value lives, you're no longer building from direct contact with truth. You're building from the emotional need to not be alone.

And that is how people end up spending years on companies that were legible to the market but never alive to them.

The boom is real. The lesson most people are taking from it is false.

Yes, the AI boom is real.

Yes, these companies are huge for a reason.

Yes, some markets really do centralize before they fragment.

None of that is the problem.

The problem is what founders do with those facts.

If you look at a month where 83% of the money went to three companies and conclude that your job is to move closer to the crowd, you've already given up the one advantage available to a founder who is not one of those three companies.

Independent sight.

Your edge is not consensus. Consensus is what you face.

Your edge is seeing something load-bearing before the crowd has language for it. Naming a problem more precisely than bigger players can. Building where actual human behavior is changing faster than the capital stack can process. Refusing to let the market's current worship pattern become your private theology.

That is the real test in an era like this.

Not whether you can raise near the center.

Whether you can stay clear enough to build from truth while everyone around you is getting high on proximity.

Because the crowd always feels strongest right before it becomes a trap.

And the founder who cannot stand apart from it will eventually confuse borrowed momentum for his own conviction.

That's the death blow.

Not being underfunded.

Not being early.

Not being ignored.

Building a future you don't even believe in because too many other people clapped when you described it.

So don't ask whether the boom is real.

Ask the harder question.

If 83% of the money is already moving in one direction, why the hell are you assuming your truth lives there too?