The AI Bubble

What is an AI bubble? Why a "bubble"?

You might have heard people talking about the "AI bubble" if you've been following the news lately. People have all sorts of opinions about it. Now, a bubble (in terms of finance) is actually quite a sophisticated topic. If I had to explain it properly, it would probably take an entire blog on its own, so I think I'll leave that for another day.

Instead, I'd rather explain what an AI bubble is with an analogy.

Take a balloon (or a bubble). Now, start filling it with air. It'll expand. Keep filling it. Every week, every day, every minute, even every second. Just keep pumping more and more air into it.

Eventually, it'll burst.

Well, that's pretty much the idea.

Now let's apply that to AI. AI companies are pouring an immense amount of capital into building AI. They're "inflating" the "AI bubble." They're spending billions, burning through funding like it's a hobby. But that's not really the main point.

Where's the actual profit?

This is where it gets interesting.

The honest answer? For most AI companies, there isn't much profit yet. Tech giants are dropping tens of billions on infrastructure, but what they're getting back is tiny compared to what they're spending. They're betting everything on the idea that AI will eventually blow up and make it all worth it.

Now, don't get me wrong, Microsoft, Google, Amazon, Meta? They're already making billions. They're not struggling. But the question investors are actually asking is different: will all this AI spending pay off enough to justify the cost today?

That's the real gamble right now.

Companies are betting AI becomes massive in everyday life. If they're right, this looks genius in ten years. If they're wrong, or if it takes way longer than expected? Yeah, the market's gonna have a rough time.

Have we seen this before?

Absolutely. Bubbles like this have been around forever. Let me tell you about two of the biggest ones.

First, the Dot-com Bubble, around 1999 to 2000. The internet was the new thing, right? It felt like the future showed up early, and everyone wanted in. Investors threw money at internet companies. Telecom companies spent billions on fiber-optic cables because they thought demand would explode.

Companies built like crazy. It felt inevitable. Everyone thought the internet would change everything overnight.

Then it didn't.

Turns out a lot of companies had no idea what they were doing. They overestimated how fast demand would grow, built way more than they needed, and a bunch of them had zero business model. When investors got scared, stock prices crashed, companies went under, and people lost their savings. Just like that.

But here's another one that's kind of wild: Railway Mania in 19th-century Britain.

Railways were brand new back then. Everyone genuinely believed trains were the future. Investors went crazy, companies raised massive amounts of money, and railways got built everywhere. The same pattern, companies built way more infrastructure than the country actually needed.Demand was expected to explode almost immediately. It eventually did, but not nearly as fast as companies had hoped.

Many railway companies couldn't make enough money to justify what they'd spent. Some went bankrupt. Fortunes disappeared. The market corrected itself.

See the pattern? First Railway Mania in the 1800s. Then Dot-com in the 2000s. And now... AI.

But here's what people don't usually talk about...

A bubble isn't always bad

This is the part that changed how I think about this stuff.

After the Dot-com crash, the fiber-optic cables didn't just vanish. They're still there, underground. And because so many had been built, companies that came later got to use them for way cheaper. That extra infrastructure that looked like a mistake? It became one of the reasons the internet exploded in growth.

Same thing happened with railways. After the bubble burst and companies failed, those railway lines didn't disappear. They became the backbone of Britain's rail network. Millions of people have used those same lines for generations.

The pattern's simple: we build too much, too early. The market corrects itself, weak companies die, prices drop, and eventually the infrastructure gets used the way people originally hoped. The "waste" becomes the foundation.

So where's AI right now?

Honestly, it looks like we're in that correction phase right now.

A lot of enterprise companies and investors thought AI would be way bigger, way sooner. Some thought it'd replace almost every job. Others believed it would take over entire industries in just a few years.

It didn't.

AI has changed how people work, sure. Some tasks got automated. But it hasn't replaced humans at the scale people expected. Turns out AI is really useful, but it's also... kind of limited.

Some companies got ahead of themselves. They laid off customer support staff because they thought AI could handle it all. Then it couldn't. AI hallucinated, misunderstood problems, kept giving the same answers. Humans could've solved it in five minutes.

So yeah, companies realized replacing humans entirely wasn't the brilliant move they thought it was.

I actually use AI almost every day and think it's amazing tech. But I'm also realistic, it's not as good as people expected it to be. Not yet, anyway. Maybe someday. Maybe not. But what history shows us is that new tech usually goes through this exact thing: huge hype, massive investment, then reality kicks in.

So... is AI the future or just hype?

I think it's the future.

We've seen this before. During Railway Mania, people said trains were the future. They weren't wrong, trains are everywhere now. Millions of people use them every single day. That bubble built something real.

Same with the internet. Around 2000, people said the internet would change everything. They weren't wrong either. Now we use it constantly without even thinking about it. Messages, videos, classes, food delivery, paying bills, it's just... normal.

I think AI's on the same path.

Right now it still feels like this exciting new technology everyone's talking about. Companies are racing to build better models, investors are throwing money at it, and people are debating if it'll actually change the world.

I think it will. Maybe not in the dramatic way people imagined a couple years ago, but I think AI will quietly become part of everything we use. Like electricity. Like the internet. Just... everywhere, but so normal we stop noticing.

And honestly? It's already happening. AI's in your phone camera, your navigation, search engines, translation apps, hospitals, factories, even the military. It's already woven into everyday life.

A decade or two from now, we probably won't even call it "AI" anymore. It'll just be... how things work. A technology in the background doing its thing, the way the internet does now.

And if history repeats, which it seems to, the AI bubble might not be remembered as the time AI failed. It might be remembered as the time we built the future a little too early.

That's kind of cool, actually.

Anyways, Hope it made some sense. Next time :D