On March 27, 2000, something remarkable happened on Wall Street.
A company that most Americans had probably never heard of became the most valuable publicly traded company in the world.
Not General Electric.
Not Microsoft.
Cisco Systems.
The company made the routers and networking equipment that helped connect the world to the internet.
And business was booming.
In fiscal 2000, Cisco's revenues would grow an extraordinary 55%, reaching nearly $19 billion.
It was profitable, growing rapidly, and supplying the infrastructure for a technology that promised to change virtually everything.
Investors weren't imagining the opportunity.
They were witnessing it.
There was just one small problem.
At its peak, investors were willing to pay more than $200 for every dollar Cisco earned annually.
By late March, the company's market capitalization had reached an astonishing $555 billion.
What happened next would become one of the most expensive lessons in stock market history.
The Future Arrived. The Returns Didn't.
The internet revolution was real.
Shopping, banking, advertising, entertainment, communication. Nearly every aspect of modern life would eventually be transformed by the technology Cisco was helping to build.
But the stock market had gotten a little ahead of itself.
In 1999 alone, the Nasdaq Composite surged nearly 86%.
Then, after peaking in March 2000, it lost approximately 78% over the following two and a half years.
Cisco's shares fell roughly 90%.
And the Nasdaq wouldn't regain its previous closing high until April 2015.
Fifteen years.
Meanwhile, the internet flourished.
Google became a household name. Amazon transformed retail. Smartphones eventually put the internet in billions of pockets.
The technology delivered.
The investors who paid the highest prices? Many had a very different experience.
Cisco wasn't a bad company. It was an extraordinary company whose stock had become extraordinarily expensive.
Haven't We Seen This Movie Before?
Fast-forward 26 years.
Another revolutionary technology is capturing the world's imagination.
Another company supplying the infrastructure has become one of the most valuable businesses on Earth.
And once again, investors are asking whether the future has changed forever.
This time, the technology is artificial intelligence.
And the company is Nvidia.
The parallels with Cisco are almost impossible to ignore.
Cisco helped build the infrastructure for the internet.
Nvidia is helping build the infrastructure for AI.
Both experienced extraordinary demand. Both became stock market giants. And both inspired investors to imagine an entirely different future.
But here's where the comparison takes an interesting turn.
The numbers tell a rather different story.
At its peak, Cisco traded at more than 200 times trailing earnings.
Nvidia, despite its extraordinary growth, has recently traded around 29 times trailing earnings.
And when we look at expected future earnings, the difference becomes even more interesting.
Recent estimates have placed Nvidia's forward price-to-earnings ratio in roughly the high teens to mid-20s, depending on the earnings period being measured.
That's a far cry from the valuations investors were willing to pay for Cisco in 2000.
For perspective, Nvidia reported approximately $216 billion in fiscal 2026 revenue, up 65% from the previous year, along with more than $120 billion in net income.
These aren't simply promises of profits that might materialize someday.
They're actual profits being generated today.
Does that mean Nvidia is inexpensive? Not necessarily.
Does it mean today's AI enthusiasm can't become another speculative bubble? Absolutely not.
But it does mean that comparing today's market with the dot-com era requires more than pointing to similar headlines.
And there's another wrinkle.
Forward earnings are estimates, not guarantees.
A stock trading at 20 times expected earnings can suddenly become much more expensive if those earnings fail to materialize.
Who Gets Rich From the Revolution?
During the internet boom, telecommunications companies spent enormous sums building networks to accommodate an explosion in anticipated demand.
Much of that infrastructure eventually became essential to the modern economy.
But not every company financing the expansion earned an attractive return.
Some went bankrupt. Others spent years recovering from investments that proved far less profitable than anticipated.
Today, technology giants are committing hundreds of billions of dollars to AI infrastructure.
Data centers. Advanced chips. Electricity. Cooling systems. Computing capacity on an almost unimaginable scale.
The demand is real. The ultimate returns remain uncertain.
It brings to mind an old investing analogy.
During a gold rush, sometimes the best business is selling picks and shovels.
But even that strategy raises questions.
What happens when customers have purchased all the equipment they need?
What happens when spending slows, competition intensifies, or the next technological breakthrough changes what equipment is required?
Nvidia has significant technological advantages, an established software ecosystem, and extraordinary profitability. It's hardly an interchangeable supplier of generic equipment.
But even exceptional suppliers ultimately depend on their customers earning enough from their investments to sustain demand.
And that's the larger question.
Will the extraordinary amounts being invested in AI ultimately generate the returns investors are expecting?
We don't know.
And neither did investors financing the internet's infrastructure in 2000.
The Trap on Both Sides
Today, some investors are convinced artificial intelligence represents an opportunity so extraordinary that traditional valuation concerns no longer apply.
Others see the similarities to 1999 and believe another collapse is inevitable.
One group sees the future.
The other sees the past.
And both may be overlooking something important.
The possibility that a compelling story is shaping their expectations more than the evidence itself.
The optimists may be underestimating the risks.
The skeptics may be underestimating how different today's financial fundamentals really are.
And both may be forgetting that two seemingly contradictory things can be true simultaneously.
The internet really did change the world.
The dot-com bubble really did happen.
A technological revolution and an investment bubble are not mutually exclusive.
Artificial intelligence may ultimately justify much of today's enthusiasm. Or investors may discover that even extraordinary businesses can become disappointing investments when expectations get too far ahead of reality.
Perhaps we'll experience some combination of the two.
History offers perspective, not a crystal ball.
The Price of Being Right
Twenty-six years ago, Cisco investors correctly recognized that the internet would transform the world.
They just couldn't know how much of that future was already reflected in the price they were paying.
Today, investors face a remarkably familiar question.
Not simply whether artificial intelligence will change the world.
But what price makes sense to pay for a future that hasn't happened yet?
Because being right about a revolution doesn't guarantee being right about an investment.
The technology changes.
The companies change.
The stories change.
Markets evolve. Human nature doesn't.
And sometimes, the most expensive mistake isn't getting the future wrong.
It's paying too much to be right.