The AI boom is everywhere—and so are the warnings. Comparing today’s artificial intelligence investment surge to the dotcom crash of 2000 has become trendy in financial circles. But here’s the thing: the AI bubble fears might be missing the forest for the trees. The infrastructure driving today’s AI economy is fundamentally different from the reckless telecom overbuilding of the late ’90s. ChatGPT hit 1 billion monthly active users in four years. The early internet took roughly the same timeframe to reach just 70 million users around 2000. That’s not a sign of instability—it’s evidence of real, scaled adoption.
The Dotcom Blueprint: A Cautionary Tale That Doesn’t Quite Fit
The late 1990s telecom sector built massive fibre-optic networks almost on faith. Operators installed thousands of miles of cable with barely 3% utilisation rates. They borrowed heavily to fund the expansion, betting revenues would eventually catch up. Spoiler: they didn’t. The whole thing collapsed, taking shareholder value with it.
Compare that to today’s AI infrastructure spending. Companies like OpenAI and Anthropic are generating over $20 billion in annual recurring revenue. Hyperscalers—think Amazon, Google, Microsoft—aren’t just hoping to monetise AI. They’re already doing it, successfully, at scale.
Why Overbuilding Is Harder Now (And That Matters)
Here’s a constraint the ’90s lacked: physics. Data centres demand enormous amounts of power and specialised cooling. You can’t just build them anywhere, and you certainly can’t build them cheaply on speculation.
That’s why hyperscalers commit customers before construction even starts. Over 90% of new data centre capacity is already allocated before the first brick is laid. Compare that to ’90s fibre networks, which were erected with vague promises and wishful thinking.
In 2026 alone, hyperscalers expect to spend over $700 billion on infrastructure. That’s real money backing real demand, not speculative excess.
The SaaS Correction: A Different Kind of Bubble Burst
Software stocks did take a hit during the so-called “SaaSpocalypse.” Public software revenue multiples dropped to around 4x—near multi-year lows. Some investors panicked. But here’s what actually happened: the market corrected overvalued growth stories, not the entire sector.
The AI situation is different. Capital Economics points out the real vulnerability isn’t valuation bubbles—it’s earnings sustainability. Investment is concentrated in a small group of highly profitable, global-scale companies. If enterprise spending on AI suddenly slows, that concentrated exposure creates genuine risk.
But “concentrated in winners” is not the same as “widespread collapse.” And it’s certainly not dotcom territory.
What Actually Separates Today’s AI Economy From 2000
The pattern differences are stark. Profitability now versus bankruptcy then: early internet companies burned money by design, while AI leaders are printing cash. Committed demand versus hopeful speculation: data centre capacity is pre-sold, fibre networks were built on faith. Measurable adoption versus hype: 1 billion ChatGPT users in four years is material adoption. And a smaller, tighter group versus startup massacre: in 2000, thousands of startups crashed; today, a handful of hyperscalers and AI firms dominate and profit.
The Real Risk Isn’t a Bubble—It’s Concentration
If there’s a legitimate worry, it’s this: the AI economy is consolidating around a few mega-profitable players. If those players’ growth slows—whether due to competition, regulation, or market saturation—the correction could be sharp. But that’s not a bubble bursting. That’s a market repricing a smaller group of winners.
Want more like this? Sign up to The MJBurrows Briefing—our free weekly newsletter delivered every Monday morning.
FAQ
Q: Is AI really more profitable than dotcom startups were?
A: Yes. OpenAI and Anthropic are generating over $20 billion in annual recurring revenue right now. Most dotcom startups had minimal revenue and burned through cash. The comparison isn’t close.
Q: Could AI spending suddenly dry up like telecom investment did?
A: Possibly, but it’s less likely. Unlike ’90s fibre networks, data centre capacity is pre-committed by customers before construction. That creates real accountability. Still, a sharp slowdown in enterprise spending would hurt concentrated AI investments.
Q: Are software company valuations in a bubble right now?
A: Software multiples have already corrected significantly—down to around 4x revenue. Whether that’s fair value depends on earnings growth going forward. But the sector survived the correction. That’s not dotcom-level collapse.
Q: What’s the biggest actual risk in the AI economy today?
A: Concentration. A handful of hyperscalers and AI firms control most of the profitable market. If their growth stalls, the correction could be sharp. But it’s not a systemic industry collapse like 2000.
Q: How many users did ChatGPT gain compared to early internet adoption?
A: ChatGPT hit 1 billion monthly active users in four years. The early internet reached roughly 70 million users over a similar timeframe around 2000. That’s 14x faster adoption, suggesting real demand, not hype.
DISCLAIMER
Effective Date: 15th July 2025
The information provided on this website is for informational and educational purposes only and reflects the personal opinions of the author(s). It is not intended as financial, investment, tax, or legal advice.
We are not certified financial advisers. None of the content on this website constitutes a recommendation to buy, sell, or hold any financial product, asset, or service. You should not rely on any information provided here to make financial decisions.
We strongly recommend that you:
- Conduct your own research and due diligence
- Consult with a qualified financial adviser or professional before making any investment or financial decisions
While we strive to ensure that all information is accurate and up to date, we make no guarantees about the completeness, reliability, or suitability of any content on this site.
By using this website, you acknowledge and agree that we are not responsible for any financial loss, damage, or decisions made based on the content presented.
