Every market boom has its poster children — the flashy, overhyped, overvalued companies that define the mania on the way up, only to crash under the weight of their own expectations. Think Pets.com in the Dot-Com era or meme stocks during the pandemic. So, the natural question is: Is the same thing happening with AI today?
Even Sam Altman — CEO of OpenAI and one of the central figures of the AI revolution — recently compared today’s AI frenzy to 1999, when internet stocks were priced for perfection.
And sure, when you look at AI darlings like Palantir trading at a PE ratio north of 500, it’s easy to see why some fear we’re in bubble territory.
But here’s what most people miss:
This isn’t just another tech bubble. It’s a generational investment cycle — and it’s just getting started.
Yes, the hype is real. But so is the spending. And that’s what separates short-term mania from long-term opportunity.
Follow the Real Money in AI
As I wrote in August’s issue of The South African Investor, the smartest way to separate hype from substance in tech is simple: follow the money.
Let’s look at the numbers.
Back in the early 2010s, Amazon, Microsoft, Alphabet, Oracle, and Meta spent less than $15 billion a year combined on infrastructure.
By 2021, that number had climbed to just under $100 billion.
Before ChatGPT launched in late 2022, it was around $150 billion.
Today? Those same five giants are expected to spend over $400 billion on AI infrastructure in the next 12 months.
That’s not a bump. It’s a supercycle.
Why This AI Spending Cycle Is Different
We’ve seen massive tech cycles before — the PC boom, the rise of the internet, the mobile era. Each reshaped the economy.
But this AI-driven cycle is different in three important ways:
1. Scale
This isn’t just more money — it’s unprecedented capital deployment.
The companies leading this charge — Microsoft, Amazon, Alphabet, Meta, Nvidia — are the largest profit engines in human history. Their capital expenditures are now measured in terms that rival the GDPs of small nations.
And because they generate enormous cash flow, they can sustain this spending for years. This gives the current AI cycle serious staying power.
2. Focus
In past tech cycles, R&D budgets were spread across dozens of experimental ideas. Many failed. A few became hits.
This time, nearly every incremental dollar is going to one thing: AI infrastructure — data centres, chips, networking, cooling, and the power to run it all.
It’s a laser-focused investment surge. That clarity creates massive opportunities for investors — especially in companies enabling this buildout.
3. The Flywheel Effect
AI infrastructure doesn’t sit idle.
Every new chip, every new data centre accelerates the pace of AI development. Faster models lead to better products. Those products generate more revenue. That revenue funds even more infrastructure.
It’s a self-reinforcing loop — a flywheel of growth that speeds up with every turn.
What This Means for Investors
This isn’t just another moment of hype. It’s a structural shift in how the world’s most powerful companies allocate capital.
And that’s why the most compelling investment opportunities in AI aren’t in the headline-grabbing apps or overhyped startups — they’re in the infrastructure behind it all.
From semiconductors to power supply to advanced cooling systems, there are companies poised to ride this $400 billion wave for years to come.
To see where we believe the best opportunities are right now, make sure you’re subscribed to The South African Investor.
Not a subscriber to Money Morning?
You can get free daily recommendations like these with Money Morning eletter. Just sign up here.
