Nvidia's New Role: Banker to Its Own AI Boom

Nvidia's record revenue masks growing role as lender, with $105B OpenAI guarantee and $200B credit exposure by 2028, raising sustainability concerns.

There's a question that tends to surface at the peak of every technology cycle, and it's not the one executives want to answer: who is actually paying for all this? For Nvidia, that question has moved from theoretical to urgent over the past week, as the company's latest earnings report collided with a series of revelations about its deepening involvement in financing its own customers.

The headline numbers were, by any measure, exceptional. Revenue for the fiscal second quarter reached $96.2 billion, up 106 percent year over year, with the data center segment surging 117 percent to $89 billion. It marked the company's 13th consecutive record quarter. The stock initially responded as expected, jumping 8.7 percent on August 27 and dragging the Nasdaq and S&P 500 higher with it, according to Reuters.

But that enthusiasm has since cooled. The shares now trade at €187.78, roughly 4.1 percent below the post-earnings peak and 7.3 percent off the 52-week high of €202.50 set in May. The stock remains 34 percent above its low from September last year, but the gap between operational strength and market reaction is becoming harder to ignore.

The Lending Machine

The most striking development isn't in the income statement — it's in the footnotes. Reuters reported that Nvidia has committed to guaranteeing up to $105 billion for OpenAI's Ohio data center lease, one of the largest AI financing commitments the company has ever made. That sits alongside a broader effort: Nvidia helped organize $500 billion in financing from six major U.S. financial institutions for customers building AI infrastructure.

The scale of this involvement raises an uncomfortable possibility. When a chipmaker promises 70 percent growth for fiscal 2028, as Nvidia did in its guidance, it may need to finance the very demand that underpins that forecast. The question investors are now wrestling with is how much of the reported boom is organic — and how much is demand Nvidia is effectively creating through its own balance sheet.

Morgan Stanley estimates Nvidia's credit exposure could approach $200 billion by the end of 2028, with roughly $170 billion of that off-balance-sheet. That structure works beautifully in an upcycle, but it would become a serious liability in a downturn.

Signs of Recalibration

The same week brought quieter signals that even Nvidia recognizes the complexity of its financing arrangements. Reuters reported that the company has paused some revenue-sharing agreements under its AI cloud financing initiative, following earlier reports that Nvidia had pulled back from the program. The company insists the business model remains intact and is simply evolving to meet demand, but the timing is notable.

There's also the pricing question. Bloomberg reported, and Reuters picked up, that customers have been informed of price increases exceeding 15 percent on servers equipped with Nvidia AI chips, affecting systems scheduled for delivery early next year, including Vera Rubin and Grace Blackwell configurations. The report remains unconfirmed, but it fits a pattern: prices rising while the financing required to support buyers becomes increasingly elaborate.

The Margin Squeeze

The cost side of the ledger is where the pressure shows most clearly. CFO Colette Kress announced that gross margins will decline to 71 to 72 percent over the next two quarters, down from 75 percent, driven by rising memory costs. Server prices may climb by up to 15 percent starting in 2027, according to reports, which could help offset some of that pressure — or simply pass it along to already-stretched customers.

The competitive landscape adds another layer. Google, Amazon, Microsoft, Meta and OpenAI are all developing their own AI chips, and OpenAI's "Jalapeno" chip has reportedly outperformed Nvidia's in testing. The U.S. government is also weighing expanded semiconductor tariffs that could eliminate existing exemptions for data centers, though that measure remains unconfirmed and unapproved.

What the Market Is Watching

Mirae Asset Securities notes that analyst focus has shifted from raw GPU demand to metrics like customer AI returns, margin stability and free cash flow. The logic is straightforward: if customers like Meta see advertising clicks improve through AI models but can't translate massive data center investments into corresponding returns, orders could eventually slow despite currently full order books.

The bull case rests on the sheer scale of committed demand. Supply obligations jumped from $119 billion to $279 billion in just three months. Amazon has reportedly ordered an additional two million GPUs for 2027 and 2028. Nvidia's financing platform, with partners including Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR, has mobilized more than $500 billion for AI infrastructure. Kress says Nvidia-financed AI labs will contribute roughly a quarter of fiscal 2027 revenue.

The proposed $12.9 billion acquisition of Hugging Face would further cement Nvidia's position in the AI ecosystem, assuming it closes as reported. CEO Jensen Huang points to the Vera Rubin architecture with its custom CPU as evidence that the company's advantage is shifting from individual chips to complete systems.

The Bottom Line

Nvidia has revealed more about its role as lender to its own industry in the past week than in many quarters combined. The 70 percent growth target for 2028 is ambitious, and the market's initial reaction suggests skepticism about whether it's achievable without increasingly creative financial engineering. The stock's annualized volatility of 45 percent reflects that uncertainty.

The next test comes with the fiscal third quarter, where Nvidia has guided to $108 billion in revenue, plus or minus two percent. Hitting that number would restore confidence in the growth narrative. Missing it, or revising margin guidance downward, would send a clear warning signal not just about Nvidia but about the entire AI sector's valuation.

For now, the shares trade above all major moving averages, and the supply commitments keep growing. But the question that opened this week's earnings story — who pays for the boom — has a more complicated answer than it did a year ago. Investors would be wise to watch Nvidia's balance sheet as closely as its income statement.