Nvidia's $4 trillion AI projection points to $20 trillion market cap

Jensen Huang's $4 trillion data center forecast implies Nvidia's market cap could quintuple to $20 trillion.

Nvidia Chief Executive Officer Jensen Huang predicted global data center capital expenditures could reach $4 trillion by 2030, a projection that implies the chipmaker's market capitalization could quintuple to $20 trillion from roughly $5 trillion today.

"The buildout of AI factories, the largest infrastructure expansion in human history, is accelerating at extraordinary speed," Huang said on Nvidia's most recent earnings call.

The math behind the projection is straightforward. The four largest AI hyperscalers — Amazon, Alphabet, Microsoft and Meta — are expected to spend roughly $650 billion on data center capex in 2026, a figure that rises to about $800 billion when including neoclouds and international players. A fivefold increase to $4 trillion by 2030 would mean Nvidia needs only to quadruple its market cap to reach $20 trillion, implying it could lose market share and still hit the target.

Nvidia already commands the vast majority of the AI GPU market, and its dominance is compounding. In its fiscal first quarter ended May 20, the company reported net income of $58.32 billion, up 211% from a year earlier, on revenue of $81.61 billion. Data center revenue alone reached $75 billion, up 92%. The stock trades at 23 times forward earnings, with 48 of 49 analysts rating it a buy and a consensus price target of $301.62, implying roughly 56% upside from its current level near $193.

The $4 trillion math checks out

The data center buildout underpinning Huang's projection shows no signs of slowing. Nvidia's supply commitments total $119 billion, underwriting the ramp of its Blackwell 300 architecture and the newly announced Vera Rubin platform. The company raised its quarterly dividend 25-fold to $0.25 per share and authorized $80 billion in additional buybacks, returning roughly $20 billion to shareholders in the most recent quarter alone.

The hyperscalers funding this expansion are deepening their commitments. Amazon spent $44.2 billion on capex in the first quarter, while Meta guided to $125 billion to $145 billion for the full year. Alphabet's first-quarter capex hit $35.7 billion, up 107% year over year. These figures support Huang's view that the infrastructure cycle has years of runway.

Who wins when models get cheaper

A separate but reinforcing thesis comes from Gavin Baker, chief investment officer at Atreides Management, who argues that as AI inference workloads shift from expensive proprietary models toward cheaper open-source alternatives, more margin dollars flow to infrastructure providers. Baker identified Nvidia, Micron Technology and Broadcom as the primary beneficiaries of this redistribution.

The numbers support the thesis. Nvidia, Micron, Broadcom and Applied Materials are projected to generate a combined $430 billion in free cash flow over the next 12 months, more than triple what they produced two years ago, according to Bank of America. Meanwhile, the five largest hyperscalers — Amazon, Alphabet, Meta, Microsoft and Oracle — are expected to see combined free cash flow fall from roughly $250 billion in 2024 to about $100 billion by the end of 2026 as they pour a projected $1.8 trillion into AI capex.

For investors, the question is whether the market has already priced in this trajectory. Nvidia's forward P/E of 23 times is well below its five-year average of roughly 40 times, suggesting earnings have outpaced the stock price. If Huang's $4 trillion projection proves conservative — as his past forecasts have — the current valuation could look inexpensive in hindsight. The next test comes when Nvidia reports fiscal second-quarter results, with management guiding for $91 billion in revenue.

This article is for informational purposes only and does not constitute investment advice.