Nvidia has struck a partnership with a group of major Wall Street financial firms to help raise $500bn (£370bn) in capital for building out artificial intelligence infrastructure, the chipmaker confirmed. The firms involved—Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR—are for the first time treating AI computing hardware and infrastructure, often called "compute," as a distinct asset class in its own right. Nvidia chief executive Jensen Huang said the funding will flow into both Nvidia's own projects and those developed by its partners.
"In AI, compute is revenue," Jensen Huang, chief executive of Nvidia, said. "We are bringing the world's leading long-term capital providers together to independently underwrite AI infrastructure."
What will the money actually build?
The bulk of the financing is earmarked for new data centres designed to house, run and cool vast arrays of stacked computer chips that process AI workloads. Alongside this, the fund will also support new factories dedicated to manufacturing the AI chips required to keep these systems running.
Huang framed the move as an evolution beyond Nvidia's origins as a chip supplier.
"Today, we are helping create a new class of productive, investable infrastructure: AI factories," he said.
Who else is backing the plan, and how firm is it?
KKR co-chief executives Joe Bae and Scott Nuttall described compute as having become indispensable to modern economic activity.
"Compute has become a critical infrastructure asset," Joe Bae and Scott Nuttall, co-chief executives of KKR, said in a joint statement. "As we've scaled our approach to digital infrastructure, we've learned that delivery, not ambition, is the hard part."
Jim Zelter, president of Apollo, which manages more than $800 million in assets, echoed that sentiment.
"Modern compute has emerged as a scarce, mission-critical asset class," he said, adding that it is also "positioned to drive significant long-term economic growth and productivity gains."
According to MarketScreener, the arrangement remains at the discussion stage, with the precise financial terms not yet fully disclosed. Separately, a Wall Street financial newspaper's report indicated the BlackRock stake in the deal runs through its Global Infrastructure Partners unit rather than the parent asset manager directly. Reporting from a business broadcaster's coverage also framed the initiative as a way to support hyperscalers, frontier AI labs and enterprise customers seeking to expand their data centres and buy Nvidia hardware.
Why does Wall Street want a piece of this?
Essentially every major technology and AI company relies on Nvidia's graphics processing units, or GPUs, to power its services, AI platforms and chatbots. Firms using Nvidia's chips include Google, Meta, Amazon, Microsoft, SpaceX, Tesla, OpenAI and Anthropic, and together they have spent more than $1 trillion on AI projects and infrastructure in just three years, with far greater spending anticipated ahead. That surging demand has driven Nvidia's stock market value up fivefold over the same period, an increase that closely tracks the company's own record $81.6bn quarterly revenue reported in May 2026.
Not every recent quarter has produced the same market reaction, however. Despite posting record revenue of $68.1bn, Nvidia's shares barely moved in February 2026 amid investor unease over AI valuations, and by late July 2026 chipmaker shares fell sharply as questions grew over the sustainability of AI spending and returns. BlackRock chief executive Larry Fink had separately warned in March 2026 that the AI boom risked widening wealth inequality, benefiting a narrow group of firms and investors.
How does this connect to Nvidia's other recent deals?
The $500bn financing plan follows a string of other large Nvidia commitments. In late July 2026, the company was reportedly in talks over a separate $250 billion guarantee tied to OpenAI's planned data-centre lease, according to a financial news agency's report. The same reporting noted that Nvidia and SK Group were also planning to build more than two gigawatts of AI data centres on the Korean Peninsula, and that Nvidia had agreed to invest $1 billion in Naver Corp to help finance an AI data centre under construction in South Korea, according to further reporting from the same outlet.
Elsewhere in the industry, BlackRock last month agreed a separate deal with Meta to finance and take a majority ownership stake in one data centre in Texas, while Anthropic recently arranged financing with Macquarie Asset Management and GIC, an investment bank in Singapore, to expand its own AI infrastructure. Anthropic did not disclose the size of that deal but said additional financing was necessary because demand for its Claude chatbot had grown so large that it now requires significant new computing capacity.
What happens next?
Nvidia has also said separately that it intends to invest up to $500 billion in AI infrastructure across the United States over the next four years, a plan that includes two new supercomputer manufacturing plants in Texas, according to a financial news outlet's report. The same report said the Texas sites are expected to begin mass production within 12 to 15 months, a timeline that will be closely watched as an early test of how quickly Nvidia can turn this fresh wave of financing into operating capacity.
Key Facts
- Nvidia has partnered with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR to raise $500bn (£370bn) for AI infrastructure.
- The financing marks the first time these firms have treated AI compute as a distinct investable asset class.
- Funds will support new data centres and chip-manufacturing factories for Nvidia and its partners.
- Nvidia separately plans up to $500bn in US AI infrastructure investment over four years, including two Texas supercomputer plants expected to reach mass production within 12 to 15 months.
- The deal follows other major Nvidia commitments, including a reported $250bn OpenAI data-centre guarantee and a $1bn investment in South Korea's Naver Corp.
