I Traded the Dot-Com Bubble and This AI Frenzy Is Different

Market participants have been fascinated this past week by a divergence that does not appear to be slowing. The VanEck Semiconductor ETF (SMH) jumped around 4.5% to a new all-time high and is grossly extended. At the same time, breadth was only about 46% positive on Friday, and the Invesco S&P 500 Equal Weight ETF (RSP) inched up about 0.3% and has not yet exceeded the highs it hit at the end of February. The bubbly moves are occurring along with very mediocre action in much of the market.

Pockets of strength and parabolic moves are not unusual. They occur in all market environments. What is different this time is that most of the stocks flying are AI-related, which has sparked comparisons to the frenzy of the 1999 to 2000 bubble.

The 2000 Comparison

CNBC recently quoated Michael Burry, famous for shorting the housing bubble in 2007 and 2008, who said the market now feels like "the last months of the 1999 to 2000 bubble." I am seeing similar chatter as people look at stocks like Intel Corporation (INTC) , Micron Technology Inc. (MU) and Advanced Micro Devices Inc. (AMD) , which led the Nasdaq 100 (QQQ) on Friday with substantial gains.

The surface comparison is easy to make. A small group of technology names is parabolic, the rest of the market is going sideways or down, and the conversation is dominated by a single thematic story. That is the 2000 setup at a high level. It is not the only era that fits the description, but it is the one most market participants are familiar with.

Burry deserves respect for the housing call. His record since then has been more mixed, and he has been calling tops with regularity for several years. The historical analogy is worth weighing on its own merits rather than on the basis of who is making it.

Why This Feels Different to Me

I actively traded the internet bubble in 1999 and 2000, and this feels quite different. The action back then was far more frantic and manic. Retail traders were convinced that the time was really different, and stocks that had no fundamentals and were valued based on pageviews could go up forever. I recall Jim Cramer writing about the "Winners of the New World," none of which survived for very long.

The FOMO of that period was palpable. Cab drivers and hairdressers had hot tips. People were quitting jobs to day trade. The phrase "this time is different" was a punchline because so many people believed it sincerely. Buyers were not thinking about exits because they could not imagine needing them.

We do not have that now. The folks chasing the chip names this week know the moves are stretched. The chatter at every gathering and on every market podcast is about how this is a bubble and that it will not last. The buyers chasing Advanced Micro Devices Inc. and Micron Technology Inc. are not euphoric. They are anxious, and most of them are planning to hit the eject button at the first sign of slowing momentum. That is the opposite of 2000 sentiment, and it has different implications for how the move ends.

Another issue worth understanding is what the money is actually chasing this time. The 1999 capex cycle was largely software, advertising and business models with no defensible economics. Pets.com and Webvan had no earnings and no path to earnings.

The current cycle is funding hard assets like data centers, advanced packaging facilities, memory fabs, power generation and grid infrastructure. The Magnificent Seven is committing roughly $725 billion in capex for 2026, with multi-year contracts behind it. Corning Incorporated (GLW) just announced a tenfold increase in U.S. optical connectivity manufacturing capacity in partnership with Nvidia Corporation (NVDA) . Micron Technology has HBM capacity sold out through 2026 against multi-year purchase commitments.

The companies leading this rally have earnings. They have visible order books. They have customer commitments stretching years out. That does not mean the stocks cannot correct. It means the structural underpinning is fundamentally different from the dot-com setup. Any correction will likely occur by time and consolidation rather than the kind of crash that follows a sentiment-driven cycle with no earnings to support it.

What Could Trigger a Reversal

We may be close to hitting a crescendo in the chip names as emotions are becoming quite extreme. We just need something to trigger it. A deal with Iran would be a perfect sell-the-news opportunity, with the market having priced that outcome over the past several weeks. The April jobs report could do it on a hot or cold print. The Nvidia Corporation report later this month is the last big earnings catalyst, and the setup into that report matters more than the report itself.

What is not going to trigger it is a Burry tweet or a CNBC segment about valuation. The bears have been calling for a top in this group for two months, and the chip names have continued higher. Tops happen on news that surprises the buyers, not on commentary that confirms what the bears already think.

My Game Plan

Regardless of which catalyst triggers the eventual rest, watch for rotational action when it comes. That is where the next significant opportunity resides. The names in the second tier, the ones that have not participated, the ones at fresh 52-week lows, are the inventory of opportunity for whatever comes next. The chip parabola eventually consolidates. When it does, the question is whether capital flows into the laggards or out of the market entirely. That is the question I have been discussing all week, and the answer comes from the equal-weight S&P 500 (RSP) chart, not from the headlines.

Have a great weekend. I'll see you on Monday.

At the time of publication, AUTHOR had no positions in any securities mentioned.