Remember March 2000? The Nasdaq was touching the sky, pets.com was going to be the future of retail, and anyone with “.com” in their name could raise a billion dollars before lunch. Then—poof. Over the next two and a half years the index got absolutely annihilated, down almost 80%. It took fifteen painful years just to get back to breakeven.
Fast-forward to today and a lot of seasoned investors are getting that same uneasy feeling in their gut. Only this time the magic words aren’t “internet” or “dot-com.” They’re “AI.” And according to one of the most watched voices on Wall Street, the poster child for everything that could go wrong is none other than OpenAI.
A Familiar Kind of Madness
Jim Cramer didn’t mince words recently. He looked straight into the camera and said OpenAI is basically “2000 in a nutshell.” That’s not a compliment. That’s a five-alarm warning from someone who lived through the last tech mania and still has the scars to prove it.
What he’s seeing isn’t just enthusiasm. It’s the same cocktail of hype, leverage, and wishful thinking that blew up two decades ago. Massive capital burns, sky-high valuations detached from revenue, and an entire ecosystem of companies whose fortunes rise or fall on one private company’s success. Sound familiar?
The difference this time? The stakes feel even bigger. Back in 2000 we were betting on websites that delivered dog food. Today we’re betting on technology that could genuinely reshape civilization. That makes the eventual reality check potentially far more violent.
Why OpenAI Specifically Triggers Alarm Bells
Let’s be honest—most private companies don’t move public markets the way OpenAI does. When their CFO makes an offhand comment about “backstopping” partners, bond traders at Oracle suddenly see their credit spreads blow out. That’s not normal. That’s the kind of ripple effect usually reserved for systemically important banks, not a startup that, technically, still hasn’t figured out a profitable business model at scale.
“They may be reckless… big swings with other people’s money.”
– Jim Cramer
That phrase should chill any investor who remembers 1999. Because reckless swings with OPM (other people’s money) was exactly how we got Webvan, Pets.com, and dozens of other flameouts that wiped out trillions in market cap.
And here’s the part that keeps some of us up at night: the entire AI trade is incredibly concentrated. Nvidia’s GPUs, Microsoft’s cloud exposure, data center REITs, uranium and nuclear names, even random companies that own land near power plants—pretty much everything traces back to the assumption that demand for frontier models will keep growing exponentially forever.
If OpenAI hits a wall—whether it’s regulatory, technical, or simply running out of cash faster than it can monetize—the dominoes could fall in every direction at once.
The Michael Burry Echo
It’s worth noting Cramer isn’t alone. Michael Burry—the guy who made a fortune betting against housing in 2008—recently warned that the market could crack before the actual spending on AI infrastructure slows down. In other words, the euphoria could deflate while the build-out is still in full swing. That’s actually scarier than the other way around.
Think about it: in 2000 the crash came after the fiber optic networks were overbuilt and the offices were empty. If Burry is right, we might not even get the satisfaction of seeing the physical infrastructure finished before the financial one collapses.
What’s Different This Time (And What Isn’t)
Every cycle, someone says “this time is different.” Sometimes they’re right. Usually they’re wrong.
Things that are different:
- The underlying technology actually works. Generative AI is real and already creating value.
- Mega-cap tech companies have real earnings and fortress balance sheets—not the case in 1999.
- Interest rates came from 1% to 5%, not from 6% to 1%. The starting point matters.
Things that feel eerily the same:
- Private valuations that make no mathematical sense.
- Public companies pivoting to “we’re an AI play now” and seeing their stocks double overnight.
- An explosion of leveraged crypto schemes masquerading as AI infrastructure (Bitcoin mining in repurposed data centers, anyone?).
- Narrative-driven pricing completely detached from current cash flows.
In my experience, when the second list starts looking longer than the first, it’s time to get cautious.
The Weird Corners of the Market Telling the Story
One of the more surreal moments last month was watching nuclear energy stocks rip higher because… ChatGPT needs electricity? Suddenly every small-cap uranium name and micro-reactor concept was “AI-adjacent.” Same with companies that own substations, cooling-tower manufacturers, even natural gas producers who swore they were yesterday’s news.
Then there were the Bitcoin miners pivoting to “high-performance computing” for AI. Same power-guzzling warehouses, same debt structure, just a new story to tell investors. If you squint, it looks a lot like 1999 companies adding “.com” to their name and watching the stock triple.
And don’t get me started on quantum computing companies promising to make GPUs obsolete. Some of these firms have market caps in the tens of billions and exactly zero dollars of revenue. That’s not innovation—that’s speculation on steroids.
So Is This THE Top?
Honestly? I have no idea. Nobody does. Markets can stay irrational a lot longer than most of us can stay solvent.
But here’s what I do know: when prominent, battle-tested investors start invoking 2000 unprompted, it’s worth listening. When the fate of hundreds of billions in market cap hinges on the success of one private company burning cash at historic rates, red flags should go up. And when random uranium stocks and Bitcoin miners become “AI plays,” we’re deep into the late innings of narrative-driven excess.
Cramer isn’t pounding the table for a crash tomorrow. He’s just saying: proceed with eyes wide open.
Maybe AI really is the transformative technology of our lifetime and all of this gets justified in the end. Or maybe we’re building the equivalent of fiber optic networks to nowhere while telling ourselves it’s different this time.
Either way, history suggests the truth usually reveals itself in the most inconvenient way possible.
So keep your powder dry, diversify away from pure story stocks, and remember: the market doesn’t care what you believe the future will look like. It only cares what actually happens.
Because if OpenAI really is 2000 in a nutshell, the ending has already been written. We just don’t know which chapter we’re in yet.
