Nvidia's Circular-Financing Debate Intensifies as August 26 Earnings Approach

Nvidia's AI growth faces scrutiny over circular financing, insider selling, and a $250B OpenAI deal ahead of Q2 earnings.

The countdown to Nvidia's fiscal second-quarter report has become a battlefield of competing narratives. On one side sits a market that has pushed the stock nearly 10 percent higher in a week; on the other, a growing chorus of skeptics — led by famed investor Michael Burry — questioning whether the chipmaker's explosive growth is built on a self-reinforcing loop of financing that one critic called "circular spending of biblical proportions."

Burry's critique, leveled just days before the August 26 earnings release, targets the intricate web of cross-investments between Nvidia and its own AI customers. The concern: Nvidia sells chips, then helps finance the very buyers of those chips, who in turn use that capital to purchase more Nvidia hardware. It's a feedback loop that flatters revenue figures while raising uncomfortable questions about the quality of demand underneath.

A Partnership Spree That Fuels the Skeptics

The timing of Burry's warning was hardly coincidental. Late July brought a flurry of announcements that gave his thesis fresh ammunition. On July 29, Nvidia unveiled a long-term strategic partnership with Safe Superintelligence, the AI startup founded by Ilya Sutskever, committing a "substantial investment" that Bloomberg pegged at $5 billion — among the largest financing deals of the chipmaker's AI boom. In exchange, the startup gains access to Nvidia's upcoming Vera Rubin GPU platform, promising a tenfold increase in compute capacity.

Days earlier, Nvidia and South Korea's SK Group had announced a partnership exceeding $500 billion to build AI infrastructure. Under the agreement, SK Telecom will construct a so-called AI factory with two gigawatts of capacity, powered by Nvidia's Vera Rubin platform, with SK hynix co-developing next-generation HBM memory chips. The first facility is slated to come online in 2027.

The pattern is unmistakable: Nvidia is simultaneously the industry's dominant supplier and an increasingly active financier of its own demand base. That dual role sits at the heart of the circular-financing critique.

The OpenAI Talks That Moved Markets

The debate sharpened considerably on July 31, when reports emerged that Nvidia was in discussions with OpenAI over a $250 billion guarantee — a deal that would see Nvidia back the AI company's ability to lease compute capacity for a US data center project. The news reignited concerns about circular financing in dramatic fashion: the stock fell roughly five percent that day, dragging the broader chip sector down with it. In a symbolic blow, Nvidia also ceded its crown as the world's most valuable publicly traded company to Apple on that same session.

Burry's subsequent warning pointed to a potential demand risk of $250 billion — precisely the scale of the OpenAI negotiations. The symmetry was not lost on market observers.

Insider Selling and Uneven Institutional Positioning

Adding another layer of complexity, media reports indicate that Nvidia insiders have sold approximately $410 million worth of shares over the past three months while the stock traded near elevated levels. Executives diversify their holdings for many reasons, and the sales alone prove little — but combined with the financing-structure questions, they form a quieter second question mark behind the growth narrative.

Institutional investors, meanwhile, are sending mixed signals. Cathie Wood's ARK funds have increased their Nvidia position while trimming Palantir, and Ken Fisher's Fisher Asset Management has reportedly added to its stake. On the other side, Suncoast Equity and Progressive Investment Management have pared back their holdings, though in relatively modest amounts. A unified institutional view is conspicuously absent.

China: The Lingering Risk Factor

Geographic exposure remains a persistent overhang. According to the risk factors listed in Nvidia's SEC filings, the company continues to face tightened US export controls on advanced AI chips, constraining sales in China's data center market. A late-June report suggested that revenue from China in the data center segment has effectively fallen to zero — a year earlier, the country had contributed billions. Media reports indicate that up to $5 billion in China-related revenue could be affected by a single policy shift.

What the Market Expects on August 26

The earnings report, covering the quarter that ended in late July, arrives alongside Nvidia's annual shareholder meeting on the same day. Consensus estimates from analysts surveyed point to earnings per share of approximately $2.13 on revenue of about $93.61 billion — though the secondary source's figure of roughly $91 billion suggests some dispersion in expectations. Wall Street projects year-over-year revenue growth of around 96 percent for the quarter, with the following quarter expected to show roughly 81 percent growth. These are analyst forecasts, not company guidance.

CEO Jensen Huang has sought to preempt concerns about production delays, reiterating that the Vera Rubin architecture is already "in production," with "enormous volumes" of additional manufacturing in the pipeline.

A Stock That Has Shrugged Off the Skeptics

The share price tells its own story. Currently trading around €190.66, the stock is up 9.50 percent over seven days — a robust recovery from recent weakness. It remains 5.85 percent below its 52-week high of €202.50, reached in mid-May, with a market capitalization of roughly €4.603 trillion.

Despite Burry's warning, 36 analysts currently count the stock among their buy recommendations. The gap between euphoria and skepticism has rarely been narrower — and the August 26 numbers will likely determine which side has been reading the situation correctly. The question is no longer simply whether Nvidia can deliver. It's whether the customers financing all those data centers can do the same.