Nvidia behind $50bn lease on Texas data centre that will use its chips
CEO Jensen Huang deploys balance sheet to backstop growth of AI computing market
Nvidia has signed leases worth up to $50bn for a massive Texas data centre, a previously undisclosed commitment that shines a new spotlight on the chip group’s growing role in financing AI.
The nearly $5tn company is leasing the entire 1 gigawatt facility that developer Hut 8 is building, which will house hundreds of thousands of Nvidia’s graphics processing units, said five people familiar with the deal.
The move is the latest example of Nvidia’s chief executive Jensen Huang aggressively using the company’s financial strength to keep it at the centre of the fast-growing market for AI computing power.
These efforts have included spending billions of dollars to foster a new generation of AI infrastructure providers, such as CoreWeave, to buy and run its GPUs. The Texas lease goes further, putting Nvidia behind the facilities that will house its chips.
The Texas site has secured access to electricity, something that is increasingly rare as developers compete for grid power. Nvidia wielded its financial muscle to lock in the site for its own chips, said an executive familiar with the deal.
“They have the balance sheet to acquire power, and in doing so, ensure their product is deployed,” the person said, asking not to be named.
Once completed, Nvidia could sublease capacity to its “neocloud” partners that buy its GPUs and sell AI cloud computing, the person said.
The arrangement will intensify concerns about circular financing, as the chip group underwrites more of the market for its chips.
Nvidia is also in talks to provide a $250bn backstop to OpenAI to help it lease a 10GW data centre SoftBank is building in Ohio, according to people familiar with the move. Details of Nvidia’s involvement in the Ohio project were first reported by The Information and The Wall Street Journal.
Nvidia’s 15-year lease commitment for the Texas site is worth $19.6bn, and renewal options would take the total value to $50bn over 30 years, according to Hut 8.
The chipmaker’s leases are central to Hut 8’s ability to cheaply finance the project, with about $4.3bn in bonds issued at 6.129 per cent in June to support the first phase of the project. The bonds drew an investment-grade rating from Moody’s because of the lease and were priced at roughly 100 basis points above the chip group’s own 30-year debt.
Hut 8 is expected to seek further financing to support a second phase of the project which will double the campus’s size and which was announced last week.
Nvidia is facing increasing competition from Google as the search giant also uses its balance sheet to push its tensor processing unit chips, including backstopping data centres and unlocking cheaper debt for developers. When a Big Tech group guarantees the data centre leases, the cost of financing roughly halves, according to bankers.
Hut 8 has publicly described its tenant at the Beacon Point project only as “a high-investment-grade company” with bond documents identifying six Big Tech companies including Nvidia as permitted tenants. The people familiar with the matter said Nvidia was the data centre’s tenant.
Nvidia’s most recent quarterly filing shows $32.4bn of data centre lease payments that have yet to commence, primarily to support research and development, up from $7.4bn a year earlier. The $9.8bn lease on the first phase of Beacon Point, which was signed in March, would account for nearly one-third of that figure.
Hut 8 said the company would not “comment on rumours or speculation regarding tenant identity” and that it has not provided guidance regarding additional financing beyond its public disclosures.
The company has said the data centre would be built around Nvidia’s DSX architecture, which standardises a site’s computing power, networking, power and cooling around the chip group’s hardware and software.
Asked about the Texas lease, Nvidia said it was “working with ecosystem partners to accelerate the deployment of efficient AI infrastructure through the DSX AI factory architecture”, which is a “blueprint for building and operating AI factories”.