Nvidia's $500B AI Compute Financing Push Widens Gap With Crypto Networks

Nvidia (NVDA) said Monday it has signed memorandums of understanding with six Wall Street firms, Apollo Global Management, Blackstone, BlackRock, Brookfield Asset Management, Goldman Sachs and KKR, to set up financing platforms that could eventually tap more than $500 billion in third-party capital. The chipmaker's goal is to treat AI compute as a bankable infrastructure asset rather than a pure tech expense, encouraging customers to build out AI data centers and lock in demand for its hardware.

"This is really the first time that technology chips have become an investable asset class. These are revenue-generating assets now. They're productive, they're long-lived, they're fungible, they're flexible," Jensen Huang, Nvidia's founder and CEO, said.

The move marks a shift in how companies pay for the raw processing power used to train and run AI models. Most businesses currently treat buying or renting computing power as a straightforward technology expense that sits on the balance sheet and loses value quickly as newer chips arrive. Nvidia's argument is that its systems are widely adopted and can be used by several customers, generating an income stream over years, so the AI factory should be treated as a long-term investable asset.

Under the MoUs, the Wall Street banks will independently assess each project for customer demand, expected utilization and cash flow before deploying capital. In some deals, Nvidia may cover 25% of the risk if the chips lose value, but the lenders still do their own checks and decide on each project themselves.

Nvidia's initiative could open up a larger pool of long-term capital at a time when investors have begun to question whether the massive capital spending by big tech firms on AI will deliver returns.

"Every industrial revolution has been built on infrastructure: electricity, transportation, communications and computing, with every buildout enabled by external financing. AI factories are the infrastructure of the intelligence era," Huang said.

The Decentralized Compute Gap

While Nvidia has become a centralized AI powerhouse, networks such as Akash and Render have looked to create a global marketplace for computing power run by ordinary people and coordinated by blockchain. They have grown, yet they have not matched Nvidia's scale.

Epoch AI finds that the largest active decentralized training networks still deliver only about one-three-hundredth the throughput of frontier data centres, noting that "it's unlikely that decentralized developers will amass frontier amounts of compute this decade." Low internet bandwidth forces GPUs to spend most of their time waiting for data rather than computing, while the need for cryptographic verification of every result imposes a heavy extra cost.

Industry analyses also flag the absence of corporate-grade service-level agreements and the practical difficulty of moving massive datasets to scattered machines as major barriers to enterprise adoption. These physical, economic and operational constraints explain why decentralized networks lag the likes of Nvidia, and the gap is set to widen with the chip maker's MoU with Wall Street banks.

Nvidia's push into infrastructure financing comes as its stock sits at $217.55, down 2.86% on the day, with an Alpha Score of 75/100. The company's NVDA stock page tracks the ongoing move. Goldman Sachs and KKR, both partners in the financing platforms, carry Alpha Scores of 46 and 47 respectively, with pages at GS stock page and KKR stock page.