Nvidia and financial giants assemble $500bn AI funding package

Nvidia is working with a consortium of major financial institutions to assemble a $500 billion (£370 billion) funding package for artificial intelligence infrastructure projects, the company said. The group includes Apollo Global, Blackstone, BlackRock, Brookfield Asset Management, Goldman Sachs and KKR, according to the announcement. The scale of the proposed package places it among the largest private-capital efforts yet aimed at the physical build-out of AI, covering data centres, power systems and the computing capacity needed to train and run advanced models.

The involvement of some of the world’s largest asset managers and private-equity firms signals that AI infrastructure is moving from a niche technology investment into a mainstream financing category. It also shows how Nvidia, whose chips are central to the AI boom, is trying to mobilise outside capital to support demand for its products without carrying the full financing burden on its own balance sheet.

Key facts at a glance

• Nvidia is working with Apollo Global, Blackstone, BlackRock, Brookfield Asset Management, Goldman Sachs and KKR.
• The target is a $500 billion (£370 billion) fund for AI infrastructure projects.
• Nvidia shares fell more than 2% on Monday after the plan was initially reported.
• Investors have raised concerns about circular financial deals in which Nvidia helps finance purchases of its own products.
• Nvidia has reportedly been in talks to backstop as much as $250 billion to help OpenAI lease computing power from a 10-gigawatt data centre that SoftBank subsidiary SB Energy is building in Ohio.
• The chipmaker is also reportedly discussing financing $350 billion of OpenAI’s purchases of its chips for the Ohio complex.

Why AI infrastructure needs huge capital

Artificial intelligence workloads require enormous amounts of computing power. Training frontier models involves thousands of specialised chips running for weeks or months, while serving those models to millions of users requires continuous access to high-performance servers, networking equipment, memory and storage. The hardware itself is only part of the cost. Data centres must be designed with sufficient power delivery, cooling systems, security and land, and they must be connected to reliable electricity grids.

Chips from Nvidia and AMD make up the majority of the cost of AI-focused data centres. That concentration means financing arrangements for AI infrastructure often revolve around the purchase of advanced processors. A large data centre can cost billions of dollars before a single model is trained, and the largest projects can require multi-year construction schedules and complex power contracts.

A 10-gigawatt data centre, such as the Ohio project described in reports, illustrates the scale involved. One gigawatt is one billion watts, so 10 gigawatts represents a very large power load, comparable to the output of multiple large power plants. Supplying that much electricity reliably requires long-term planning with utilities, grid operators and local authorities. It also raises questions about energy availability, environmental impact and the pace at which new transmission and generation capacity can be added.

Circular financing and concentrated risk

The announcement comes at a sensitive moment for AI-related stocks. In recent months, investors have expressed jitters around hundreds of billions of dollars in circular financial deals, in which companies such as Nvidia have helped finance the purchase of their own products. Such arrangements can raise concerns about concentrated risk because the same companies that supply hardware may also provide or backstop the capital used to buy it.

Vendor financing is not new in the technology industry. Equipment suppliers have long offered credit or leasing arrangements to help customers afford large purchases. In the AI sector, however, the sums involved are unusually large and the number of creditworthy buyers is relatively small. When a chipmaker helps finance a customer’s purchase, the supplier’s fortunes become more closely tied to the customer’s ability to generate returns from those chips. If AI demand slows or projects are delayed, the financial exposure can become more visible.

Nvidia shares fell more than 2% during the day on Monday after the news was initially reported. The decline suggests that investors are weighing the benefits of a large funding package against the risks of deeper financial entanglement across the AI supply chain. On one hand, the consortium could unlock capital for projects that might otherwise struggle to secure financing. On the other, it could increase the market’s focus on how AI spending is being funded and who ultimately bears the risk.

The Ohio data centre and OpenAI

Nvidia has reportedly been in talks to backstop as much as $250 billion to help OpenAI lease computing power from a 10-gigawatt data centre that SoftBank subsidiary SB Energy is building in Ohio. The chipmaker is also reportedly in discussions to finance $350 billion of OpenAI’s purchases of its chips for the Ohio complex. The figures are striking even by the standards of the AI boom, and they show how financing discussions can span both the physical infrastructure and the chips that will run inside it.

OpenAI is one of the most prominent buyers of AI computing capacity, and its demand for chips has become a major factor in the broader semiconductor market. A project of the Ohio complex’s size would require coordinated financing across several layers: the real estate and construction company, the energy provider, the data centre operator, the chip supplier and the ultimate customer. Each layer introduces its own risks and return expectations.

The reported discussions also highlight the role of SoftBank and its subsidiaries in AI infrastructure. SB Energy’s involvement in a 10-gigawatt project reflects the growing intersection between energy development and AI computing. Power has become a binding constraint for data centre growth in many regions, and companies that can secure large-scale electricity supplies are gaining strategic importance.

What the financial consortium brings

Apollo Global, Blackstone, BlackRock, Brookfield Asset Management, Goldman Sachs and KKR are among the largest players in private capital, infrastructure investment and asset management. Their participation could take several forms, including equity investments, private credit, infrastructure funds, lease financing and structured products. The consortium structure may allow the firms to share risk while deploying capital at a scale that would be difficult for any single institution.

Asset managers have been expanding their infrastructure portfolios as institutional investors seek long-duration assets with inflation-linked cash flows. Data centres and related power infrastructure fit that profile, but they also carry technology risk. The useful life of AI hardware can be short relative to traditional infrastructure, and the pace of change in chip design can affect the value of assets used as collateral. Lenders and investors must therefore assess not only the credit quality of the operator but also the durability of the underlying technology.

Goldman Sachs and KKR bring experience in large corporate transactions and private equity, while BlackRock and Blackstone are major allocators of institutional capital. Brookfield has a long history in infrastructure and renewable power. Apollo has been building a large credit and insurance-linked investment platform. A combination of these capabilities could support projects from early construction through to operation, but it could also create complex interlocking relationships among suppliers, customers and financiers.

Market reaction and the scrutiny ahead

The initial market reaction was cautious. A decline of more than 2% in Nvidia shares is modest relative to the stock’s recent volatility, but it shows that investors are sensitive to any arrangement that might blur the line between hardware sales and financial engineering. Analysts have been debating whether AI spending is generating enough end-user revenue to justify the scale of investment in data centres and chips.

That debate has intensified as more deals involve suppliers, customers and investors with overlapping interests. Circular financing can make demand look stronger in the short term because it helps customers buy products they might not otherwise afford. It can also create a feedback loop in which the supplier’s revenue growth supports its own valuation and its ability to raise capital, which in turn supports further financing for customers.

The proposed $500 billion package could help address one of the biggest bottlenecks in the AI build-out: access to capital at a scale large enough to fund multi-gigawatt projects. It could also accelerate the construction of data centres and power infrastructure that take years to plan and build. But the structure will likely face scrutiny from investors, regulators and creditors who want to understand who is ultimately responsible for repayment and how risk is distributed across the consortium.

The broader AI financing landscape

AI infrastructure is becoming a capital-intensive industry in its own right. Beyond chips, the build-out requires transformers, substations, cooling equipment, fibre networks and skilled construction labour. Supply chains for many of these components are already stretched, and lead times for critical electrical equipment have lengthened. Financing packages that include power infrastructure could therefore have an impact beyond the technology sector, affecting utilities and industrial suppliers.

The consortium approach may become a template for future projects. By bringing together several financial institutions, Nvidia and its partners can spread exposure and tap different pools of capital. However, the complexity of such arrangements can make it harder for outside investors to assess risk. Disclosure will be important, particularly if the deals involve guarantees, backstops or lease obligations that do not appear directly on a company’s balance sheet.

The reported Ohio project and the broader $500 billion package also underscore how AI demand is reshaping corporate finance. Chipmakers are no longer just selling components; they are helping to organise the capital that makes large-scale computing possible. Financial institutions are moving into AI infrastructure as an asset class, and energy companies are becoming central players in the technology supply chain. The result is a web of relationships that links semiconductor design, data centre construction, power generation and private capital.

For now, the key facts are clear: Nvidia is working with Apollo, Blackstone, BlackRock, Brookfield, Goldman Sachs and KKR on a $500 billion package; the plan is aimed at AI infrastructure; and Nvidia shares fell more than 2% after it was reported. The reported discussions with OpenAI and SB Energy show the scale of individual projects under consideration. The market’s response and the final structure of the financing will determine whether the package is seen as a necessary catalyst for AI build-out or as another sign of circular risk in the sector.

Source: Silicon UK News