My children often tell me that “today” is the best or worst day ever. Sometimes both land on the same day. Apparently it is possible to have the worst day ever multiple times a week.
Unrelated.
The benchmark S&P 500 keeps tagging fresh records and the tech-heavy Nasdaq Composite refuses to roll over. And yet Wall Street’s most famous bear has called the top again.
In early May 2026, Michael Burry told his more than 200,000 Substack subscribers that “the market has jumped the shark” and that “the end of… this… is nigh.” He has also taken a significant leveraged short position, reportedly through January 2027 put options on the semiconductor ETF SOXX.
The comparison that should give every long-only mega-cap tech investor pause is the one Burry actually made. The Philadelphia Semiconductor Index, which includes NVIDIA, Broadcom, Micron, Intel and TSMC, rose more than 10% in the single week ending May 8 and is up roughly 65% in 2026 alone. Burry argues that kind of velocity looks like the parabolic run semiconductor and tech stocks put in during the final months before the Nasdaq peaked in March 2000.
The poster child of the cohort is NVIDIA (NASDAQ:NVDA | NVDA Price Prediction), now a $5.45 trillion company carrying the AI factory build-out on its back. I have held NVIDIA for over 15 years, and the fundamentals are still genuinely impressive.
Q4 FY2026 revenue landed at $68.13B, up 73.2% YoY, with Data Center revenue of $62.31B and Data Center Networking up 263% YoY as Grace Blackwell ships. Free cash flow ran $96.58B for FY2026. Jensen Huang’s framing is that “the agentic AI inflection point has arrived.” The demand side reads like a memo from the future: OpenAI’s 10-gigawatt commitment and Anthropic’s 1 GW initial buildout, both backed by direct NVIDIA investment.
Now the valuation gut-check. NVIDIA trades at a trailing P/E of 43 and a forward P/E of 24, with a price-to-sales of 23. The wider market is the louder warning. The S&P 500 Shiller PE Ratio sits above 40 as of May 2026, the highest reading since the 2000 tech bubble and well above the historical average of about 17. Some AI firms have raised funding at up to 150 times their revenue. Goldman Sachs CEO David Solomon has acknowledged that “a bunch of the capital that’s being deployed will actually not produce any returns.”
There is a breadth problem too. Fewer than 55% of S&P 500 components trade above their 50-day moving average even after a 7% move in the index itself. Strategist Jim Paulsen has argued that rallies led by new era stocks have generally held up when the rest of the market stayed reasonably correlated, and that when those stocks race ahead alone, it has usually been a warning sign.
Burry leans on long memory here. He has compared today to four prior episodes: the 2000 dot-com crash, the 2008 Great Recession, the 1970s Middle East oil crisis, and the 1929 crash. The 2000 mirror is the closest. The Nasdaq fell roughly 78% from its March 2000 peak to October 2002, and Cisco, the networking darling of that era, lost about 89% peak-to-trough.
Burry’s operative phrasing is measured. “Reject greed.” “Consider taking profits.” And “this, all of it, is the scene of the bloody car crash, minutes before it happens.”
If you buy the bull case, you own NVIDIA and the broader top 10 because you believe the AI capex cycle compounds for another decade and that data center demand absorbs the $95.2B in supply commitments NVIDIA now carries on its books.
If you read the Shiller PE above 40, sub-55% breadth, and 150x revenue funding rounds as the same conditions investors saw in 1929, 1973, 2000 and 2008, you trim, hedge, or pair the position with cash. Long term, the U.S. equity market has always come back stronger, and NVIDIA’s one-year return of about 81% sits on a five-year return near 1,328% and a ten-year return above 20,000%. The decades take care of themselves. The next twelve months are what Burry is staring at.
For me? One day my children will be right about the best or worst day ever. One of them has to be, eventually. Right?
Top Gaining Stocks
Top Losing Stocks