NVIDIA is leveraging its formidable balance sheet to gain market influence, offering financial guarantees to emerging cloud providers in exchange for a share of their revenue, quietly evolving from a chip vendor into a central authority within the AI computing ecosystem.
According to a July 1st report by The Information, NVIDIA is providing financial backstop commitments to young cloud service companies that lease its GPUs. Should these firms fail to find enough AI developers to rent their computing power, NVIDIA will lease back their unsold GPU capacity at an agreed price.
In exchange, NVIDIA will take a percentage of these cloud service providers' revenue, with the share proportion gradually decreasing over the contract term. GPU cloud providers Firmus and Sharon AI are already participating in this program, and three executives with business ties to NVIDIA have confirmed the arrangements.
On July 1st, NVIDIA announced on its official website a new business model combining revenue sharing with credit support. This model allows AI cloud vendors to procure NVIDIA infrastructure without fully bearing the upfront capital expenditure themselves, enabling them to provide computing services to downstream AI-native companies, model developers, and enterprise clients.
Shifting Business Model
The report indicates this initiative is referred to internally by some at NVIDIA as the "AI Compute Partnership," a fact confirmed by a company spokesperson. This move marks a significant strategic pivot for NVIDIA.
On one hand, it aims to expand its customer base by lowering the financing barrier for emerging cloud providers. On the other, it allows NVIDIA to directly participate in the profit distribution of the downstream computing market through revenue sharing, extending its control over the AI supply chain further downstream.
From Chip Sales to Cloud Revenue Sharing
According to NVIDIA's official news release, the company will receive standard product revenue plus an additional share of cloud service revenue, creating a recurring income stream tied to usage. The core intent of this model is to dismantle the financing barriers that have long constrained startup AI companies from accessing large-scale computing power.
NVIDIA positions this framework as the "DSX AI Factory" model, targeting AI service scenarios requiring cross-region continuous operation, high utilization, and multi-tenant accelerated computing.
Sharon AI and Firmus are the first cloud providers to adopt this model. Sharon AI plans to deploy up to 40,000 NVIDIA Grace Blackwell GB300 GPUs. Firmus is building a DSX AI Factory campus on Batam Island, Indonesia, expected to scale to 360 megawatts and host up to 170,000 NVIDIA GPUs. These deployments directly illustrate NVIDIA's latest progress in translating computing demand into deployable, financeable infrastructure.
NVIDIA notes that emerging AI companies have historically faced severe constraints in accessing capital-intensive infrastructure, as even long-term commitment contracts were often insufficient to secure financing for computing purchases. This meant many AI-native companies, model developers, and inference service providers faced lengthy waits when scaling compute capacity, with each step—site selection, power procurement, construction, hardware commissioning—potentially taking months or longer.
The new model promises to allow these entities to access full-stack accelerated computing capabilities more rapidly, bypassing the traditional infrastructure build cycle, through an economic realignment.
The Guarantee Mechanism
The report states that GPUs are typically the most expensive component in an AI data center. For chip buyers with lower credit ratings, obtaining sufficient loans is itself a major hurdle.
Commenting on such transactions, one data center executive described them as "killing two birds with one stone." He explained that if NVIDIA only backed the lease for the data center facility, the problem of "how to finance the GPUs" would remain. But if NVIDIA promises to pay for unsold computing capacity within the facility, "you've solved the GPU financing problem, and you've solved the data center financing problem."
In essence, NVIDIA's guarantee acts as a credit enhancement tool, enabling emerging cloud providers who would otherwise struggle for bank loans to leverage larger-scale capital and accelerate data center construction.
Strategic Motivation
NVIDIA's rollout of these measures has a clear strategic context. Currently, a handful of large cloud providers—including Amazon, Microsoft, SpaceX, Oracle, Meta, and Google—purchase the majority of NVIDIA's chip output. However, several of these companies are developing competitive AI chips in-house, posing a potential threat to NVIDIA.
To reduce dependence on these giant customers, NVIDIA has, over the past few years, consistently supported a cohort of emerging GPU cloud providers, exemplified by CoreWeave. The "AI Compute Partnership" is a continuation and deepening of this strategy.
According to a previous report by The Information, NVIDIA has also recently been in talks to provide financial guarantees for a large data center OpenAI is leasing in Ohio. If built to full scale at current prices for chips, labor, power, and other materials, the cost could reach $500 billion.
Financial Commitments
NVIDIA's financial commitments in this direction are already substantial.
To date, NVIDIA has invested billions of dollars in several emerging cloud providers in exchange for equity and, in some cases, agreed to lease back chips from these companies, involving firms like CoreWeave and Lambda in deals totaling billions. A previous The Information report noted that NVIDIA's own researchers have used GPU servers leased back from Lambda.
Regarding capacity guarantees, NVIDIA began advancing such deals last fall. In September 2024, NVIDIA committed to purchasing all of CoreWeave's unsold capacity through 2032 if it couldn't find tenants, in a contract valued at $6.3 billion at the time. This effectively alleviated investor concerns about CoreWeave's highly leveraged business model, driving its share price up nearly 30% in the following week.
According to a regulatory filing NVIDIA submitted in May of this year (covering the quarter ending in April), the company has since added another $3.5 billion to guarantee customer data center leases in exchange for the right to purchase their stock.
Overall, NVIDIA is constructing a multi-layered mechanism of interest alignment: equity investment, capacity leasebacks, lease guarantees, and now revenue sharing. Each layer deepens the financial ties between NVIDIA and downstream cloud providers, enabling NVIDIA to directly share in the incremental profits from AI computing commercialization beyond mere chip sales.
Comments