CoreWeave Chief Executive Michael Intrator said Nvidia is “covering their bases” as the chipmaker moves to make sure its customers have enough computing power through a widening network of AI infrastructure deals. His comments came as Nvidia’s market value climbed above $5.30 trillion, a staggering rise from about $750 billion in May 2023.
Intrator’s remarks landed after Nvidia reached an agreement with IREN to work on installing about 5 gigawatts of DSX-branded AI infrastructure across IREN’s data center platform. As part of that arrangement, Nvidia gained the right to buy up to 30 million IREN shares at $70 apiece over five years, a potential stake worth $2.1 billion. For a company that already dominates the AI chip market, the message is clear: Nvidia is not only selling the hardware that powers artificial intelligence, it is also helping shape the supply chain that keeps that hardware in use.
The scale of Nvidia’s rise explains why that strategy is drawing attention. Its market capitalization has increased more than sevenfold since May 2023, and demand for its AI chips has been described as having increased more than sevenfold over the same period. Nvidia’s stock has risen 16.33 percent over the last month and 88.16 percent over the previous year. CoreWeave’s stock rose 12.45 percent over the last month and 123.28 percent over the previous year, underscoring how closely investors continue to track the broader AI buildout around Nvidia.
Intrator said the more important question is whether Nvidia can expand the AI ecosystem far enough to keep most of the demand on its own platform. That is the tension at the center of Nvidia’s current approach: it is trying to preserve its grip on AI infrastructure by pairing chip sales with partnerships that guarantee customers can actually deploy the computing capacity they need. The IREN deal suggests Nvidia is willing to spend heavily to secure that position while the market is still growing fast.
For now, the company’s dual strategy of selling chips and reinforcing the ecosystem around them appears to be working. The harder test will be whether that model can hold as more of the AI industry begins to build out its own infrastructure and compete for the same demand Nvidia has helped create.
