Sequoia Capital partner Shaun Maguire went on CNBC last week and said SpaceX (NASDAQ:SPCX), freshly trading, looks to him “more like Nvidia three years ago” than Tesla (NASDAQ:TSLA | TSLA Price Prediction). He also said he plans to hold his shares “forever.” Sequoia is a longtime SpaceX backer, so the incentive to talk his book is obvious. Still, the comparison is worth unpacking because it is a specific claim about where SpaceX sits on the curve, and the curve has a recent, very expensive precedent.
The NVIDIA comparison, and why he rejected the Tesla one
Three years ago, in June 2023, NVIDIA (NASDAQ:NVDA) traded at a split-adjusted $39.41. The AI thesis was contested, hyperscaler capex was just beginning to inflect, and bears framed the stock as a cyclical chip name riding a temporary GPU shortage. Since then, NVIDIA shares are up 419.89%, the company carries a $4.95 trillion market cap, and Q1 FY27 data center revenue alone hit $75.25 billion, up 92% year over year. CEO Jensen Huang called the buildout “the largest infrastructure expansion in human history.” You can read the underlying 8-K here.
Maguire’s framing implies SpaceX is at the analogous moment. Customers are real, the infrastructure thesis is concrete, and the multiple has not yet priced in what he thinks 2029 and 2030 revenue will look like. Tesla gets rejected because it often traded on narrative rather than on visible contractual revenue. SpaceX’s Connectivity segment generated $11.39 billion in 2025, with segment adjusted EBITDA of $7.17 billion, growing 49.8% year over year. That is the part of the business already paying for the harder parts.
The three-year growth catalysts Maguire is underwriting
He expects “dramatic growth” over the next three years from three vectors. Starship, orbital data centers, and Starlink direct-to-cell. SpaceX says Starship V3 should carry 100 metric tons to orbit, and the vehicle could eventually reduce the cost to reach orbit by 99% or more. Drop launch cost by two orders of magnitude and the addressable market reorders itself.
The orbital data center pitch is wilder. SpaceX expects to begin deploying orbital AI compute satellites as early as 2028, eventually a constellation of potentially millions of satellites running inference workloads in sun-synchronous orbit. The xAI acquisition closed in February 2026 and now forms the AI segment, which generated $818 million in revenue in the first quarter alone while burning operating cash on compute buildout.
The “hold forever” model and what’s actually behind it
Maguire said the quiet part out loud. “I have what I think the company’s revenue is going to be in 2029, 2030. And I have what I think is a reasonable multiple on that. The answer I get to is a very big number.” He also called SpaceX’s mission “the most important mission of any company in history.” That second part is venture-capital register. The first part is a DCF dressed up in conviction language.
Early backers have an obvious reason to be vocal at debut. Newly public stocks routinely sag around lock-up expiration as insiders sell. None of that invalidates the long thesis, but it shapes how a public-market investor should pace any position.
Key-man risk and the public-market workarounds
On Elon, Maguire said “Elon is the most visionary entrepreneur of all time. I also think he’s underappreciated in his operational ability.”. SpaceX’s S-1 is blunter, describing the company as “highly dependent” on Musk and noting it does not maintain key-person life insurance on him. He also runs Tesla, holds roles at Neuralink and The Boring Company, and previously served as Senior Advisor to the President.
For exposure to the same ecosystem, Tesla carries a $2 billion equity stake in SpaceX and shares Musk’s attention. Shares are down 7.2% year to date at $406, though Polymarket assigns a 90.5% probability that SpaceX carries the higher valuation between the two by June 30.
The closer launch comparable is Rocket Lab (NASDAQ:RKLB), up 34% year to date and 285% over the past year. Q1 revenue grew 63.5% to $200.35 million, backlog reached $2.2 billion, and the company was selected for the Department of War’s Space Based Interceptor program. Neutron, the medium-lift rocket meant to match Falcon 9, slipped later into 2026 after a stage-1 tank test failure. The valuation, at 102.6 times trailing sales, already prices in a lot of what has not happened yet. Which, oddly enough, is also Maguire’s argument for SpaceX. The difference being he gets to hold his shares forever at the cost basis Sequoia paid years ago, and you do not.
Top Gaining Stocks
Top Losing Stocks