The company no longer merely makes the chips of the AI boom. It also invests, commits to capacity, backs infrastructure and organizes the capital that makes buying those chips possible.
Published on August 29, 2026 · by Alex · Nvidia · Artificial intelligence · Financing
Nvidia reported quarterly revenue of $96.221 billion, up 106% from a year earlier. Data Center contributed $89 billion and set another record.
Yet the figure that best explains the company's new role is not at the top of the income statement. It is spread across investments, receivables, capacity commitments and guarantees. Nvidia is no longer just selling AI infrastructure: it is helping finance the system that buys it.
At quarter end, Nvidia held roughly $99 billion in equity investments and another $25 billion in investment commitments. The portfolio spans public and private holdings across manufacturers, infrastructure providers and the wider AI ecosystem.
Accounts receivable reached $63.059 billion. Dividing that ending balance by quarterly revenue produces roughly 60 days of sales, versus about 45 days at the previous fiscal year end. Nvidia does not report that as an official metric, but it shows that a larger share of sales had yet to turn into cash.
| Part of the system | Reported figure |
|---|---|
| Quarterly revenue | $96.221 billion |
| Data Center revenue | $89 billion |
| Accounts receivable | $63.059 billion |
| Equity investments | $99 billion |
| AI cloud commitments | $36 billion |
| Maximum guarantee exposure | $108.5 billion |
The newest mechanism appears in agreements with select AI clouds. Those companies buy Nvidia systems while Nvidia commits to purchase some of their capacity, normally over six years. The commitments totaled $36 billion.
The clouds can sell that capacity to third parties at better rates, and Nvidia may share in revenue under certain conditions. That creates a loop: the cloud buys hardware, the commitment helps the buildout, third parties consume the capacity and Nvidia can receive an additional share of the business.
Maximum exposure is not the same as a payout. Nvidia disclosed $3.5 billion of guarantees tied to select AI cloud obligations and up to $105 billion of conditional credit support for an SB Energy campus in Ohio that will host Nvidia systems under OpenAI leases.
The $108.5 billion has not been handed out. Most of the exposure would phase in from fiscal 2029, depends on data centers entering service and triggers payments only under defined defaults. It declines as OpenAI meets its lease payments.
Even with those conditions, the shift matters: Nvidia's balance sheet is backing land, power and buildings that will hold more of its own systems.
In August, Nvidia also signed memorandums with large capital providers to mobilize more than $500 billion toward AI infrastructure. As explained in our earlier analysis of those arrangements, they are preliminary: they may not become definitive agreements, and financing would be independently underwritten by capital providers rather than automatically supplied by Nvidia.
As long as demand grows, this position is extraordinarily powerful. Nvidia sells the chip, invests in the ecosystem, reserves capacity and helps new projects find financing.
The risk is circularity. How much demand would exist independently, and how much relies on commitments, guarantees and investments from the same supplier recording the sale?
The central-bank comparison is an analogy, not an accounting definition. But it identifies the right change: Nvidia's balance sheet is now part of AI infrastructure. In coming results, receivables, investments and guarantees may say more about the quality of growth than another revenue record.
Primary sources: Nvidia's fiscal 2027 second-quarter results and its Form 10-Q filed with the SEC.
Because it no longer just sells chips: it invests across the ecosystem, commits to capacity purchases, backs parts of the infrastructure and helps mobilize outside capital. It is an analogy for Nvidia’s role in the system, not a claim that the company is a bank or creates money.
No. That is maximum gross exposure: $3.5 billion for obligations of select AI clouds and up to $105 billion in conditional credit support for the SB Energy project serving OpenAI. Much of it becomes effective only as future phases come online and a default occurs.
Accounts receivable, equity investments, AI cloud commitments and guarantee exposure. If they grow faster than sales, the dependence between Nvidia and the customers buying its infrastructure will increase.
This article is the written version of the Saturday analysis.
Watch the video on TikTok · Follow the channel
