There's a lot of debate these days about whether AI is a bubble. I'm not going to argue either way. What I wanted to look at was what history says might happen if AI were a bubble that popped. If we look back at the dot-com bust, it took the Nasdaq 15 years to recover its prior peak. If history were to repeat itself, it suggests that the Invesco QQQ (NASDAQ:QQQ), an ETF that tracks the Nasdaq-100 index, wouldn't recover until 2042 if it popped within the next year.

I'm not predicting this will happen at all, as I'm bullish on AI and the Nasdaq-100. However, I still think it's a good idea to at least consider this potential scenario before allocating too much of a portfolio to one top ETF that has so much exposure to the AI megatrend.

Missed AI's "Act 1"? Act 2 Could Be 15x Bigger. Most investors think they missed the AI boat because they didn't buy Nvidia in 2005. But according to our analysts, we're only at the end of "Act 1"—the R&D phase. "Act 2" is the global rollout. Continue »

Image source: Getty Images.

Bursting the bubble

The bursting of the dot-com bubble ranks as one of the biggest stock market crashes in history. The internet-driven rise in the Nasdaq Composite index started in 1995 when it was below 1,000 points. The tech-heavy index would go on to rise to a peak of 5,048 points on March 10, 2000, a more than 400% gain in about five years. The index subsequently crashed a gut-wrenching 77% from that peak, bottoming on Oct. 4, 2002, at 1,139.90. It took the Nasdaq 15 years to recover from this crash, finally reaching its prior high on April 24, 2015.

The primary factor causing the crash was the overvalued stock market. Many investors speculated that dot-com companies would eventually be immensely profitable, even though many weren't generating any revenue at the time. In late 1999, the Nasdaq traded at a price-to-earnings ratio of more than 200. The likely catalyst triggering the crash was the Federal Reserve's decision to raise interest rates, which constrained capital flows and made it more challenging for cash-strapped internet companies to raise capital to fund their operations.

Recognizing a historical pattern

There are some eerily similar patterns developing today. The tech-heavy QQQ is up more than 90% over the past three years, driven by AI-related enthusiasm. Meanwhile, the Nasdaq-100 currently trades at nearly 34 times earnings, up from 32 times last year, and above its historical average of 22.6 times over the last two decades.

Tech companies are investing heavily in AI, increasingly funding it with debt. Over the past year, U.S. hyperscalers, including Alphabet (NASDAQ:GOOG)(NASDAQ:GOOGL), Amazon, Meta, Microsoft, and Oracle, have issued a combined $220 billion in debt to fund data center development, chip purchases, and other AI-related investments. They'll likely continue to issue debt to fund their AI build-out. That's a concern, given that the Federal Reserve recently raised interest rates for the first time in three years and plans to continue hiking them to tame inflation.