Nvidia Lines Up $500 Billion Financing Push as AI Race Moves to Credit Markets
Nvidia has assembled six major financial firms to build financing platforms capable of mobilizing more than $500 billion for AI infrastructure, signaling that access to capital is becoming the decisive factor in the global AI competition.
Nvidia has brought together six of the largest names in finance — Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR — to create financing platforms that could mobilize more than $500 billion for companies building artificial intelligence infrastructure. The move marks a shift for the chipmaker from supplying hardware to organizing the broader financial machinery behind the AI buildout.
The arrangement reflects a wider change in the industry. While public attention remains fixed on model releases, chip performance and data center capacity, the variable increasingly determining who can compete is money. The next phase of the AI race, among companies and countries alike, will depend largely on who can pay for it, and the action is moving from frontier laboratories to credit desks.
The scale of the spending explains why. The five largest American hyperscalers are on track to spend $730 billion on capital expenditure in 2026, and a shrinking share of that will come from their own balance sheets. They are expected to finance up to a third of it with debt. AI has turned technology from an asset-light business — software written once and distributed at negligible cost — into a heavy industry requiring factories, advanced chips, substations and enormous supplies of electricity.
Nvidia's answer borrows from project finance, the model long used to fund power plants, toll roads and airports, where high upfront costs are paid down by steady returns over many years. The new effort attempts to turn computing capacity into an asset class that pension funds and insurers are comfortable owning, much as they already own infrastructure. Crucially, Nvidia and its six partners will not, for the most part, be lending their own money. Capital is to come from third parties: pension funds, insurers, sovereign wealth funds and household savings.
That structure is meant to answer familiar concerns about circular financing, in which a chipmaker's money returns to it as orders. Rather than a closed loop, the buildout is being wired into the wider financial system, restructuring the cost of supply for a broader bench of competitors beyond the hyperscalers, who can themselves only partly afford the expansion.
The risks are real. The parallel with infrastructure is imperfect: energy facilities are built to last decades, usually in response to foreseeable demand, and generate predictable returns. Graphics processing units are short-lived assets, AI demand is unpredictable, and it is too early to project returns. Still, if the upside from AI is speculative, so is the downside — and if the model works, it could reshape the AI market much as venture capital changed the calculus for starting a business.
The geography of capital matters as much as its quantity. The United States starts with an edge because no other country is as good at turning the savings of pensioners and households into risky, long-lived assets at scale. America's bond markets, private credit funds and asset managers were already looking for exactly this kind of asset, which is how a chipmaker could line up a half-trillion-dollar financing target in a single announcement.
Europe illustrates the opposite problem. The continent holds one of the world's largest pools of savings, with eurozone households alone holding about 10 trillion euros, roughly $11 trillion, in bank deposits. Yet European institutional investors would rather buy U.S. tech bonds than back their own startups. The money is there; what is missing is the plumbing to turn savings into investment. If AI comes down to access to capital, Europe's AI strategy needs a capital-markets strategy, and progress on long-stalled capital-markets integration would do more for competitiveness than another round of gigafactories.
China's difficulty runs the other way. State money is good at pouring concrete but less good at matching capacity to demand; after a policy-driven building spree, Chinese media reported last year that as much as 80% of newly built computing capacity sat unused. The Stanford HAI AI Index 2026 estimates that private investment in AI in the United States last year was more than 23 times that in China.
The first phase of the AI boom relied on technical breakthroughs in models, chips and applications. The next phase depends on the financial system's ability to underwrite the industry's expansion beyond a small set of companies. Nvidia has made a bold attempt to reconfigure the economics of AI, and competitors will need equally imaginative financing to keep pace.
5
