Markets
June 1, 2026
Binance Adds 7,000 US Stocks as Crypto Exchanges Race To Become ‘Everything Apps’
Key Takeaways
- Michael Burry compared Nvidia to Cisco during the dot-com bubble, despite the firm posting $81.6 billion in quarterly revenue
- Burry warned that hyperscalers account for roughly half of Nvidia’s data center revenue, creating concentration risk if AI spending slows
- The Philadelphia Semiconductor Index is up roughly 65% in 2026 as Burry holds put options tied to a potential reversal in AI stocks
Nvidia’s stock has spent most of this year doing the kind of things that normally silence critics. Revenue jumped 85% year over year to $81.6 billion last quarter, the company approved an $80 billion buyback, and the broader SOX index is up roughly 65% in 2026 alone. Yet Michael Burry is leaning harder into the other side of the trade. In a recent Substack post, the investor behind ‘The Big Short’ compared Nvidia to Cisco during the dot-com era. He argued that the AI boom may be creating risks investors are ignoring.
Also Read: Binance Adds 7,000 US Stocks as Crypto Exchanges Race To Become ‘Everything Apps’
Why Is Michael Burry Comparing Nvidia to Cisco in 2000
Burry’s argument is less about whether artificial intelligence is useful. It is more about how the current AI infrastructure cycle is being financed. In his post, Burry pointed to hyperscalers like Microsoft, Amazon, Meta, and Google, which together account for roughly half of Nvidia’s data center revenue. This level of concentration matters because a pullback from even one large customer could ripple quickly through the supply chain. Burry wrote,
“NVIDIA is benefitting from strong demand, but is selling into a concentrated set of buyers whose own demand is being distorted by a training and benchmarking phase that will not last.”
He also highlighted what he described as a “bullwhip effect.” This is where companies over-order chips and data-center capacity during periods of extreme demand. Nvidia then locks in massive long-term supply commitments to keep up.
According to Nvidia’s filings, the company now carries more than $100 billion in future supply obligations. These are tied largely to manufacturing agreements. Burry argues those commitments were built around an AI spending cycle that may not stay this aggressive forever.
Also Read: Citi Says Tokenized Securities Could Become a $5.5 Trillion Market by 2030
AI Infrastructure Trade is Getting Bigger and More Concentrated
One reason Burry’s comments gained attention is that Nvidia’s earnings were objectively strong. The company beat Wall Street expectations again, but the stock reaction was muted compared to previous quarters.
Not everyone agrees with Burry’s Cisco comparison. Some argue Nvidia’s earnings growth still supports the rally, unlike the dot-com era, where valuations ran far ahead of actual profits.
This has led some investors to question whether expectations around AI stocks have simply become too large to satisfy consistently. Burry also pointed to a Moody’s estimate showing major hyperscalers carry roughly $662 billion in off-balance-sheet lease commitments tied to data-center expansion. Those obligations do not fully appear on balance sheets yet. But they reflect how aggressively companies are building around AI demand.
Burry said Nvidia reminds him of Cisco because both companies sat at the center of major technology buildouts. Cisco’s business remained profitable after the dot-com crash, but its stock still fell sharply once spending slowed.
Also Read: SpaceX IPO Frenzy Sends Virgin Galactic Stock Soaring as Traders Pile Into SPCE
