Nvidia Revealed as Mystery Tenant Behind Hut 8's $50 Billion Texas AI Data Center

Nvidia Revealed as Mystery Tenant Behind Hut 8's $50 Billion Texas AI Data Center

Nvidia Revealed as Mystery Tenant Behind Hut 8's $50 Billion Texas AI Data Center

Nvidia is the mystery tenant behind Hut 8's massive Beacon Point AI data center campus in Texas, according to a Financial Times report that surfaced this week. The chip giant reportedly signed two 15-year leases worth a combined base value of $19.6 billion, with renewal options that could push the total contract value past $50 billion.

The arrangement marks a sharp shift in how Nvidia approaches the cloud market. Instead of just selling GPUs, the company is now using its own balance sheet to lock down scarce power and data center capacity, then planning to sublease space to "neocloud" providers that buy its hardware.

Blade server chassis in a blue-lit data center rack

A Campus Built Around Nvidia's Own Architecture

Hut 8 has spent months quietly commercializing Beacon Point, a one-gigawatt site in Nueces County, Texas. The campus has secured 1,000 megawatts of utility capacity through an interconnection agreement with AEP Texas, and the company fully commercialized the site in July after signing a second lease worth $9.8 billion with the same tenant.

The facility is being designed around Nvidia's DSX reference architecture, which coordinates computing, networking, power, and cooling for large-scale AI factories. Hut 8 expects the site to hold hundreds of thousands of Nvidia GPUs, handling both AI training and inference workloads.

Initial energization is scheduled for the first quarter of 2027. The first data halls tied to the second phase are due for delivery in the second quarter of 2028. That timeline puts Beacon Point on the same cadence as several other mega-campuses under construction across Texas, where the state's deregulated power market and fast permitting have made it the default destination for AI build-outs.

Server rack with tangled network cabling in a data center aisle

Why a Chipmaker Is Acting Like a Landlord

The deal structure matters beyond the headline number. Nvidia's participation reportedly helped Hut 8 line up about $4.3 billion in financing for the first phase, because investment-grade lease commitments give lenders long-term contracted revenue they can underwrite against.

The Financial Times reported that Nvidia could sublease capacity to neocloud companies — the young cloud operators that buy Nvidia GPUs and resell AI computing services. Pulse 2.0's rundown of the FT story lays out the lease math in detail. That arrangement extends Nvidia's reach well beyond the processor business. The company gets to guarantee its hardware actually gets deployed, and it can offer infrastructure to customers who can't independently finance a project of this scale.

Supporters say these deals fix a real bottleneck. AI infrastructure is being held up by shortages of power, land, and financing, and a chipmaker with a strong balance sheet can break all three logjams at once. Nvidia reported more than $130 billion in cash and marketable securities at the end of its last fiscal year, giving it the firepower to underwrite projects that would strain the balance sheets of most dedicated data center developers.

The Circular Financing Question

Critics see a different problem. Nvidia would be financing infrastructure that runs its own processors, then selling access to that infrastructure to companies that buy even more Nvidia equipment. That loop has drawn comparisons to the kind of circular financing that worry investors in other boom-and-bust tech cycles.

Hut 8 estimates the first two Beacon Point phases will involve roughly $17 billion in combined investment, supporting about 1,900 construction jobs and 230 permanent operating roles. The lease value breaks down as 704 megawatts of contracted IT capacity across two 352-megawatt phases.

Neither Nvidia nor Hut 8 has publicly confirmed the tenant's identity. Yahoo Finance, carrying the Reuters report, said it could not independently verify the Financial Times story, and Nvidia declined to comment on the record. The silence is telling, though — this is the second major AI infrastructure move linked to Nvidia in as many weeks, following reports it is weighing a $250 billion guarantee to help OpenAI lease SoftBank's 10-gigawatt Ohio campus.

What It Means for the Cloud Market

If the deal closes as reported, it changes the competitive math for hyperscalers. Amazon, Microsoft, and Google have spent the past year racing to build data centers, with AWS growth hitting 37% and Azure climbing 43% in the latest quarter. Nvidia entering the capacity business adds a fourth force — one that doesn't compete on software but controls the supply of the chips everyone else needs.

For neoclouds, the deal could be a lifeline. Companies like CoreWeave and Nebius have grown fast by renting out Nvidia GPUs, but they've struggled to secure the long-term power contracts and construction financing that hyperscalers take for granted. Leasing from Nvidia-backed campuses sidesteps that problem, even if it makes those companies more dependent on the chipmaker.

There's also a question of what this does to Nvidia's own returns. The company is sitting on enormous cash flows from AI chip sales, and parking some of that in data center leases is a bet that AI demand stays strong for decades, not just quarters. Hut 8's Beacon Point timeline — first power in Q1 2027, full build-out through 2028 — shows how far out these commitments run.

The Energy Angle

Texas has become ground zero for the AI data center boom, and Beacon Point sits in the middle of that wave. The ERCOT grid has added gigawatts of new load from data centers in the past two years, and utilities like AEP Texas have had to rework interconnection queues to keep up. Nvidia's lease effectively converts a power reservation into a revenue contract — lenders can underwrite against it, and the utility can plan transmission upgrades around it.

That's part of why the deal structure is being studied so closely. Every hyperscaler has its own version of this playbook, from Amazon's direct agreements with nuclear plant operators to Google's investments in small modular reactors. Nvidia's version is different because the company is neither a cloud operator nor a utility — it's a component supplier inserting itself into the middle of the infrastructure chain.

What the Analysts Are Saying

Wall Street has spent the week parsing what the lease means for both companies. For Hut 8, the deal removes most of the risk that has historically dogged bitcoin miners pivoting to AI hosting: no speculative building, no empty racks, no reliance on spot market pricing. The company's shares have climbed since the campus was fully commercialized, with analysts noting the investment-grade tenant removes the financing discount that smaller data center developers usually pay.

For Nvidia, the questions are about capital allocation. The company has been clear that it intends to keep returning cash to shareholders while funding growth, and a multi-billion-dollar lease book is a different kind of commitment than building chips. Some analysts argue the leases are effectively a form of demand creation — Nvidia ensures its next-generation GPUs have a home before they even ship, which supports the aggressive production plans it has shared with suppliers like TSMC.

The counter-argument is that Nvidia is taking on balance-sheet risk in a market that has already shown signs of froth. Data center vacancy rates remain near record lows, but the sheer volume of announced capacity has some investors asking whether every project will find a tenant. If the AI demand curve flattens, lease obligations could become a drag on the very margins that made Nvidia the most valuable company in the world.

The Bottom Line

Nvidia's reported $50 billion commitment at Beacon Point is a bet that the AI cloud build-out has years to run. It turns the chipmaker into a landlord, a financier, and a potential competitor to the very cloud providers that buy its products.

The risks are real. A demand slowdown would leave Nvidia holding expensive leases with no tenants, and the circular-financing optics won't sit well with every investor. But for now, the company is signaling it sees the AI infrastructure shortage as the biggest constraint on its own growth — and it's willing to put its balance sheet behind fixing it.

Hut 8, meanwhile, gets what every data center developer wants: a blue-chip tenant, construction capital, and a fully subscribed campus. The market will find out in Q1 2027 whether the bet pays off, when the first phase of Beacon Point is scheduled to light up.

← Back to Home