Nvidia has suspended parts of a recently introduced financing program for artificial intelligence cloud providers, a quiet retreat that underscores how quickly the chipmaker's expanding financial footprint across the AI industry is drawing regulatory and investor scrutiny.
The company pulled back from some transactions under its AI Compute Partnership Program less than two months after the initiative was introduced, according to PYMNTS, which cited media reports and people familiar with the matter. The program offered financial support to cloud operators purchasing Nvidia hardware while giving the chipmaker a share of the revenue generated from that computing capacity. For more context on this story, see our ongoing AI industry coverage.
A Financing Model That Raised Eyebrows Internally
Under the initial proposal, Nvidia could receive 50% of revenue above a specified threshold from cloud providers operating systems built with its chips. Early participants included Sharon AI and Firmus Technologies. Nvidia also offered to lease unused GPU capacity if providers were unable to secure other customers — effectively a backstop that could make it easier to finance data-center construction.
The design was meant to solve a real problem. Smaller AI cloud companies often must commit billions of dollars to chips and data centers before they have enough customer contracts to support conventional financing. Nvidia's guarantees and credit support could reduce that risk while accelerating deployment of infrastructure built around its processors.
But the same structure put Nvidia in multiple roles within a single market: chip seller, financier, and revenue participant. According to the report, some Nvidia employees raised concerns with current and prospective customers that the arrangement could invite antitrust scrutiny, and potential partners objected to restrictions Nvidia sought to impose on how they rented out computing capacity.
Nvidia Says the Model Is Evolving, Not Dead
The precise reason for the suspension remains unclear, and the initiative could eventually be restructured or folded into other programs. Nvidia has maintained that the broader business model remains active and is evolving as demand for AI computing increases.
That distinction matters for the company's customers. The suspension covers parts of a specific program, not Nvidia's overall strategy of using its balance sheet to accelerate AI infrastructure. The company has become increasingly aggressive in this area, and the financing debate is unfolding against the backdrop of extraordinary core-business growth: Nvidia recently reported quarterly revenue of $96.22 billion, up 106% year over year, with data-center sales of $89 billion, and projected roughly $108 billion in revenue for the following quarter, per PYMNTS.
The Circular Financing Question
The pullback arrives as analysts and investors increasingly question what some describe as circular financing within the AI industry — arrangements in which the dominant chip supplier helps fund its own customers' purchases and, in some cases, participates in the revenue those purchases generate.
Nvidia's recent moves illustrate the scale of the question. The company joined Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR in an initiative targeting more than $500 billion of third-party capital for AI infrastructure, according to the report. Nvidia has also offered guarantees of as much as $105 billion connected with OpenAI's data-center leasing plans.
Each arrangement makes sense individually; together they paint a picture of a supplier whose capital is now woven into nearly every layer of AI buildout, from chips to data centers to the rent that pays for them. That is precisely the pattern antitrust regulators tend to examine, and it explains why Nvidia's own employees reportedly worried the revenue-share program crossed a line.
What It Means for the AI Infrastructure Boom
For small AI cloud operators, the suspension removes a financing path that was uniquely suited to their economics. Backstops from a vendor with Nvidia's balance sheet made it possible to borrow against future demand; without them, providers like Sharon AI and Firmus Technologies will need conventional financing — which, ironically, requires the customer contracts the program was designed to bridge.
For Nvidia, the retreat is a calculated de-escalation. The company still books record revenue from hardware sales alone, so shedding the most scrutiny-prone parts of a two-month-old program is a cheap concession if it reduces the odds of a formal antitrust probe clouding its growth story.
The ripples extend beyond Nvidia itself. TSMC, the chipmaker's primary manufacturing partner for advanced processors, is directly exposed to the trajectory of these financing arrangements: Nvidia's Blackwell processors and its forthcoming Rubin platform rely heavily on TSMC's advanced manufacturing and packaging capabilities, tying the foundry's prospects closely to the pace of AI infrastructure deployment. Any cooling in the rate at which new AI cloud operators can finance chip purchases eventually shows up in ordering patterns further down the supply chain.
Investors will be watching whether the program reappears in restructured form — and whether regulators say anything about the version that just got shelved. The suspension is the clearest signal yet that the AI industry's financing experiments are now being audited not just by markets, but by the people who write antitrust enforcement.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
