Nvidia Doubled Down on Its Second-Largest AI Holding. Should Investors Follow Suit?

Nvidia nearly doubled its CoreWeave stake in Q1 to more than 47.2 million shares, or roughly 11% of the company, valued at about $4.9 billion. CoreWeave posted Q1 revenue of $2 billion, up 112% year over year, while backlog jumped 284% to $99.4 billion and management guided to $31 billion-$35 billion of 2026 capex. The article is broadly constructive on AI infrastructure demand and CoreWeave’s growth, though profitability remains absent and valuation is still elevated.

Analysis

The market is starting to treat AI infrastructure as a land grab rather than a normal capex cycle, and that changes who captures value. CRWV is the direct beneficiary if its backlog converts cleanly, but the more durable winner may be NVDA: every incremental deployment tightens chip allocation, extends pricing power, and increases the value of being the bottleneck supplier rather than the landlord of depreciating hardware. The second-order risk for CRWV is that the market is underwriting growth on contract backlog while ignoring asset intensity and financing dependence; if debt or equity markets tighten, the equity can re-rate sharply even before operating demand slows.

The more interesting nuance is that the upside case for CRWV is not just revenue growth, but operating leverage on utilization. If deployments remain contract-profitable, the market will eventually stop valuing it like a hyper-growth software name and start valuing it like a constrained infrastructure utility with cyclical financing risk. That transition is months away, not days, so near-term price action will likely be driven by sentiment around AI spend durability and any evidence that the backlog is converting faster than capex.

Consensus may be underestimating how asymmetric this is for NVDA versus CRWV. NVDA has the better risk-adjusted exposure because it monetizes demand regardless of which neocloud wins share, while CRWV has single-name execution and balance-sheet risk layered on top of an already crowded AI trade. The main contrarian bull case for CRWV is that scarcity value in GPU capacity could keep multiples elevated longer than fundamentals would normally justify, but that also makes the stock vulnerable if investors begin to distinguish booked demand from cash flow generation.

Key Decisions for Investors

• Maintain/add to NVDA on pullbacks over the next 2-6 weeks; preferred expression is core long equity because it captures AI demand growth without CRWV’s financing and execution risk.
• Trade CRWV tactically only, not as a long-duration compounder: buy on post-event weakness for a 1-3 month momentum trade, but size small due to dilution and capex funding risk.
• Pair trade: long NVDA / short CRWV into strength if CRWV rallies on backlog headlines; this isolates hardware monetization vs. capital-intensive infrastructure risk.
• If long CRWV, consider using call spreads instead of stock for the next 3-6 months to cap downside from any funding overhang while retaining upside if AI demand remains frothy.
• Watch for any signs of capex guidance creep or debt market widening over the next quarter; those would be the earliest signal to reduce CRWV exposure and rotate toward NVDA.