Amazon is reportedly exploring a new financing structure involving approximately $8 billion worth of Nvidia's advanced artificial intelligence chips as the company continues to expand its massive data-center infrastructure.
According to a Financial Times report cited by Reuters, Amazon has been discussing the potential transaction with investors. The proposed arrangement would move thousands of Nvidia Grace Blackwell chips into a special-purpose vehicle, or SPV, while Amazon would lease the hardware back for continued use.
The reported plan comes as Amazon commits enormous amounts of capital to artificial intelligence and cloud computing infrastructure.
How the Proposed Deal Would Work
Under the reported structure, Amazon would transfer thousands of Nvidia Grace Blackwell chips into a separate investment vehicle.
The SPV would then raise money from outside investors, potentially through debt issuance. Amazon would lease the chips back from the vehicle and continue using them in its data centers.
The proposed structure would allow Amazon to shift ownership of expensive semiconductor equipment away from its own balance sheet while maintaining access to the computing hardware.
Amazon is also reportedly considering offering investors an equity stake of up to 10% in the special-purpose vehicle.
The Chips Are Already Being Used
The Nvidia chips involved in the proposed transaction are not simply future orders. According to the reports, the hardware has already been purchased or leased by Amazon and installed across more than a dozen U.S. data centers.
The facilities are located across five states, including Nevada and Virginia.
Nvidia's Grace Blackwell platform is designed for demanding artificial intelligence workloads, including the training and operation of advanced AI models.
These systems combine powerful computing processors with high-speed memory and networking technologies designed for large-scale AI infrastructure.
Why Amazon Is Exploring the Structure
Building AI data centers requires enormous amounts of capital.
Companies such as Amazon, Microsoft, Google and Meta are investing heavily in servers, specialized AI chips, networking equipment, power systems and data-center facilities as demand for AI computing continues to increase.
For Amazon, using an asset-light financing structure could provide another way to fund its expansion without keeping the full value of the semiconductor equipment on its balance sheet.
The approach is similar in concept to equipment leasing, where a company continues using critical hardware without necessarily owning all of the underlying assets.
Amazon's AI Infrastructure Spending Is Massive
Amazon has been rapidly expanding its capital spending as it builds additional computing capacity for Amazon Web Services and other AI-related workloads.
The company has indicated that its capital expenditures for 2026 are expected to exceed $200 billion, with a large portion directed toward AWS infrastructure, AI chips and new data centers.
The scale of the investment reflects the growing demand for computing power from artificial intelligence services.
AI models require large numbers of specialized processors, and cloud providers are competing to secure enough computing capacity for customers building and running AI applications.
Nvidia Chips Are Becoming Financial Assets
The proposed Amazon transaction also highlights how valuable advanced AI chips have become within the technology industry.
Modern AI accelerators are among the most expensive components inside data centers. Large clusters can contain thousands of processors, creating billions of dollars in hardware investments.
Because these chips have significant economic value, financial institutions and investors are increasingly exploring ways to use AI infrastructure as part of financing structures.
The development could create new financing models for companies building large AI computing facilities.
A New Model for Financing AI Data Centers
The potential Amazon transaction comes during a broader shift in how technology companies are thinking about financing AI infrastructure.
Hyperscale cloud companies are spending at unprecedented levels to build data centers, but maintaining rapid expansion can place pressure on corporate balance sheets and borrowing requirements.
Asset-backed financing and leasing structures can provide companies with alternative sources of capital while allowing them to continue operating the underlying infrastructure.
For investors, such structures can create exposure to physical technology assets and the cash flows generated from their use.
The Deal Has Not Been Completed
It is important to note that the reported Amazon transaction is still being discussed and has not been announced as a completed deal.
The Financial Times reported that Amazon has been speaking with potential investors to assess interest in the proposed structure.
Amazon and Nvidia had not immediately commented on the report when Reuters published its coverage.
Therefore, the final value, structure and terms of the transaction could change or the proposed financing could ultimately not proceed.
What It Means for the AI Industry
If completed, the transaction would provide another example of how the rapid growth of artificial intelligence is changing not only computing technology but also the way infrastructure is financed.
AI companies and cloud providers need increasingly large amounts of computing power, while the cost of acquiring and operating that infrastructure continues to rise.
Financial structures that allow companies to spread or transfer the cost of expensive AI hardware could become increasingly important as data-center construction accelerates.
Amazon's AI Expansion Continues
Amazon is investing across multiple areas of artificial intelligence, including cloud computing, AI model development and specialized infrastructure.
Amazon Web Services is competing for customers that need large-scale computing resources to train and deploy AI systems.
The company is also developing its own AI chips while using processors from companies such as Nvidia and other semiconductor suppliers.
The reported $8 billion Nvidia chip financing proposal therefore represents one part of a much larger infrastructure expansion.
As AI demand continues to grow, the technology industry is increasingly focused not only on building faster processors but also on finding new ways to finance the enormous infrastructure required to operate them.
Journalist: Vijay Singh