Key Highlights

  • Discussion of neoclouds often overlooks Nvidia, which has a direct stake in their success.
  • AI clouds, enterprises and other customers are now almost as large a Revenue channel for Nvidia as hyperscalers.
  • That mix gives Nvidia Leverage over hyperscalers, which cannot easily slow Nvidia purchases to ramp their own chips.
  • The argument is that customers chasing top AI performance would move to neoclouds offering the newest Nvidia chips.
  • Neocloud balance sheets and Credit markets remain the main risk to the thesis.

The Player Missing From the Neocloud Debate

Most discussions about the future of neoclouds focus on the neoclouds themselves: how they raise capital, how fast they add capacity and how they compete with the large cloud providers. One participant is often left out of that conversation, even though it sits at the center of the industry. That participant is Nvidia (NASDAQ: NVDA).

The argument is straightforward. Neoclouds are no longer a side channel for Nvidia. Together with enterprises and other customers, AI clouds have become almost as big a source of revenue as the hyperscalers. If that is right, Nvidia’s position changes, and so does the logic of how the rest of the industry behaves.

What the Numbers Show

Nvidia now reports Data Center revenue in two groups: hyperscale, and a second group covering AI clouds, industrial and enterprise customers, known as ACIE. In the latest quarter, hyperscale revenue was $48.7 billion and ACIE revenue was $40.3 billion, based on figures reported in Market Analysis of the results. Hyperscalers’ share of data center revenue slipped to 55% from 59%.

Growth is also faster outside the hyperscalers. According to analysis by Tomasz Tunguz, hyperscale revenue rose 13% from the prior quarter, while ACIE revenue rose 25% and was up 138% from a year earlier. The same analysis says neoclouds contributed the majority of net-new data center revenue for the first time.

Nvidia’s management has said that AI-native clouds do not design their own chips and do not want to. For these customers, Nvidia is the platform.

Why Hyperscalers Cannot Simply Walk Away

Hyperscalers are building custom chips, and some analysts expect Nvidia’s share of the AI accelerator market to ease from near 90% toward a range of roughly 70% to 80% by the end of 2026. But the leverage argument suggests that Nvidia’s broader customer base makes it harder for the hyperscalers to push back.

The reasoning runs like this. A hyperscaler that slowed its Nvidia purchases to develop and ramp its own chips would risk falling behind in a race where performance matters. Customers in the AI race are unlikely to settle for weaker hardware. If a hyperscaler cannot offer the newest Nvidia chips, those customers could move to neoclouds that do. Nvidia has another set of buyers ready to absorb the demand.

That is what keeps the industry on the Nvidia standard. The neoclouds act as a competitive check on the hyperscalers, and Nvidia benefits from the pressure.

Why Nvidia Wants Neoclouds to Thrive

If that logic holds, the survival of neoclouds matters directly to Nvidia, and the company has reason to do everything it can to keep them alive and thriving.

There are signs it is already doing so. Tunguz’s analysis says Nvidia has extended payment terms, with Days Sales Outstanding rising from 45 to 60 days and receivables reaching $63 billion. It also points to about $101 billion of Equity held in AI startups and neoclouds buying its chips. Because non-hyperscalers have weaker balance sheets and smaller cash flows than Big Tech, the analysis expects Nvidia to provide more supplier financing to sustain their growth.

The Risks to the Thesis

The same dependence cuts both ways. Neoclouds are exposed to credit markets, and funding costs can change quickly. Goldman Sachs recently upgraded CoreWeave (NASDAQ: CRWV) to Buy, but it also noted that the stock was down 43% since Aug 11 amid credit market volatility. That Volatility raises questions about whether, and at what rate, CoreWeave can finance its next phase of growth. Goldman still sees CoreWeave as differentiated from other neoclouds and expects its ARR per MW to be biased to the upside over the next 24 months because of tight supply.

Nvidia’s own financing exposure is another consideration. Extending credit and taking equity stakes in customers ties Nvidia’s results more closely to the health of those customers. Custom chips from hyperscalers also remain a long-term competitive factor.

What Investors Should Watch

Investors can track whether ACIE revenue keeps growing faster than hyperscale revenue, whether the hyperscale share continues to fall from 55%, and how neoclouds fund expansion. Further shifts in Nvidia’s payment terms and customer financing would also be telling. The leverage argument rests on neoclouds staying financially healthy, so their access to Capital is the variable to follow.

Disclaimer: This content is for informational purposes only and does not constitute financial advice or a recommendation to buy, sell, or hold any security. Investors should conduct their own research and consult a licensed financial adviser before making Investment decisions. Analyst estimates and company data are subject to change.