In early February, the Nasdaq suffered its worst three-day slide since April, with over $1 trillion in market value shed, driven—at least in part—by AI bubble fears. When you combine this with over 100,000 job cuts last month—the most for the first month of the year since 2009—it starts to look like the AI bubble's days are numbered. Since we've seen a tech bubble pop before (back in 2000) it's worth looking back and drawing some comparison, to see what we can learn.
Looking back, it’s easy to point to the exact moment a bubble bursts. But in 2000, the "pop" wasn’t a single event; it was a long, painful deflation. The NASDAQ peaked on March 10, 2000, but it didn't hit rock bottom until October 4, 2002—over two and a half years later—having shed nearly 78% of its value. If history repeats, we wouldn't see a true floor for 24–30 months.
Looking at jobs, in 2000 the economy was at "peak employment" with an unemployment rate of 3.9%. Jobless claims began to climb steadily through 2001, eventually peaking as the recession took hold. Today, we see similarities: unemployment has hovered around 4.0% – 4.2% (as of mid-2025/early 2026 data). While the "Magnificent Seven" have driven the index to record highs, continuing jobless claims have recently begun to hit their highest levels in years. This "divergence"—where the top 10 stocks soar while the average worker feels the squeeze—is exactly what we saw in the final innings of 1999.
In 1999, the market was led by the 'Four Horsemen': Microsoft, Intel, Cisco, and Dell. They were the picks-and-shovels of the internet era. Much like today’s AI darlings, investors believed these companies were invincible because everyone had to buy their hardware to get online. But when the bubble burst, even the 'invincible' Cisco saw its stock price crater by 80%, a level it hasn't truly eclipsed on an inflation-adjusted basis for over two decades.
No discussion of the AI bubble would be complete without a nod to the current King of Hype, Elon Musk. During recent earnings calls and public demos, Elon Musk has begun describing the combination of Tesla’s Optimus humanoid robot and Full Self-Driving (FSD) as an "infinite money glitch."
“With AI and robotics you can actually increase the global economy by a factor of 10 or maybe 100. There's not like an obvious limit. [The Telsa Robot] is kind of like an infinite money glitch and maybe there won't even be money in the future, or money, but it might be measured in terms of wattage, like how much power can you bring to bear from an electrical standpoint? So I guess what I'm saying is hang on to your Tesla stock.”
Thanks for the stock tip, Elon.
This sounds a like the predictions being made during dotcom mania. In August of 1998, Time Magazine, which held one of the largest circulations for any weekly news magazine in the world, predicted the end of retail shopping with the headline story "Kiss your mall goodbye."
Only, retail didn't go away. And except for some dips in 2008 and 2020 (for obvious reasons) retail has been on the rise from 1998 on through present day.
Retails sales in the United States from 1992 to 2025. Source.
When the tech "Four Horsemen" collapsed in 2000, capital didn't just evaporate—it moved. Investors fleeing the dotcom wreckage found shelter in three specific places:
• Value & Small Caps. While the NASDAQ was in a freefall, the Russell 2000 Value index actually rose nearly 50% between 2000 and 2002.
• The "Defensive" Fortress. Investors piled into "Old Reliable" sectors—Utilities, Healthcare (Johnson & Johnson), and Consumer Staples (Walmart).
• U.S. Treasuries: In 2000, bonds were the ultimate fire escape. As the Fed slashed rates to cushion the crash, Treasury prices soared, providing a double-digit silver lining for investors.
This is where the 2000 analogy starts to fail us. In the dot-com era, you could hide in "boring" stocks or government debt. Today, the landscape is more complex:
• The Valuation Contagion. Unlike 2000, many "defensive" stocks have already been bid up to premium prices as investors anticipated this slowdown months ago.
• The Treasury Trap. In 2000, the U.S. deficit wasn't a headline-dragger and inflation was tamed. Today, with massive deficits Treasuries may be a dangerous place to seek shelter.
• The Rotation Risk: If the AI giants stumble, the sheer size of their market cap weight means they might pull the "boring" sectors down with them through index fund liquidation—a "forced selling" phenomenon that wasn't nearly as prevalent 25 years ago.
So are we seeing the end of the AI bubble? Quite possibly. But if 2000 taught us anything, it’s that a bubble doesn't pop overnight—it deflates. We are likely in the "show me the money" phase. In 2001, the internet was real, but it took a decade for the business models to catch up to the hype.
AI is real, too. But the bridge between "cool demo" and "bottom-line profit" will almost certainly be a longer walk than the stock market currently expects.