Nvidia faces more risk from slower new agreements and tighter China scrutiny than from leadership uncertainty, Bloomberg Intelligence said, according to a Bloomberg News report on 12 November. Nvidia’s own filings give a separate measure of its investing: net additions to its private-company stakes were $13.1 billion in the quarter to 26 July 2026, down from $17.9 billion the quarter before, a 27% fall by our arithmetic.
The stakes still grew. Nvidia carried $47.9 billion of non-marketable equity securities at 26 July 2026, against $3.8 billion a year earlier, about 12.6 times as much by our arithmetic, and its supply and capacity commitments rose to $279 billion from $119 billion the quarter before, the same 10-Q says.
Business Pill · Forecast vs Commitment Explained
An evergreen Business Engineer explainer on the difference between a forecast and a commitment. It explains the general idea only and does not discuss Nvidia.
The key insight: Bloomberg Intelligence ranks slower new agreements and tighter China scrutiny as larger risks for Nvidia than leadership uncertainty, Bloomberg News reported. Nvidia’s filings show $47.9 billion of private stakes at 26 July 2026, about 12.6 times a year earlier, and quarterly net additions of $13.1 billion, down 27% from $17.9 billion, both by our arithmetic.
The opening of the Bloomberg News article, the part we could read, ranks two business risks, slower new agreements and tighter China scrutiny, above leadership uncertainty. It does not name the analysts, say which agreements it means, or give figures for either risk.
Nvidia reports its fiscal third-quarter results on 17 November 2026. The filed figures below run to 26 July 2026.
Quarterly net additions to Nvidia’s non-marketable equity securities and the carrying value at each quarter end, from Nvidia’s 10-Q and 10-K notes. The fourth-quarter fiscal 2026 figure and the 27% fall are our arithmetic.Animated: the carrying value of Nvidia’s non-marketable equity securities at each quarter end, July 2025 to July 2026, in billions of dollars, from Nvidia’s 10-Q and 10-K filings. The 12.6 times multiple and the 27% fall in quarterly net additions are our arithmetic.
The Stakes: $47.9 Billion, Up 12.6 Times
Nvidia’s 10-Q describes its non-marketable equity securities as “primarily in privately-held companies carried at cost less impairment, and adjusted for observable price changes”. The carrying value rose from $3,799 million at 27 July 2025 to $8,187 million in October 2025, $22,251 million in January 2026, $42,336 million in April 2026 and $47,898 million at 26 July 2026, the filings show.
In the July quarter, net additions were $13,106 million and unrealized gains $4,900 million, while $12,323 million was reclassified, which the filing says was “primarily related to marketable securities following public market trading”. The 10-Q says the stakes carried cumulative gross unrealized gains of $9.1 billion.
New Stakes Slowed in the Latest Quarter
Quarterly net additions went from $299 million in the quarter to July 2025 to $3,704 million to October 2025, $12,792 million to January 2026 (the 10-K’s $17,444 million for the year minus $4,652 million for nine months, our arithmetic), $17,899 million to April 2026 and $13,106 million to July 2026. By our arithmetic, the July quarter was the first decline in the series shown here.
Separately, the April 10-Q listed the investments Nvidia made in that quarter as “$18.6 billion in private companies and infrastructure funds”, a different measure from net additions, and said some include “AI model makers that may indirectly purchase or use our products in the cloud”.
Nvidia’s commitments and guarantees as listed in its 10-Q for the quarter ended 26 July 2026; the categories overlap in kind and timing and are not additive.
Commitments Beyond the Stakes
The July 10-Q says Nvidia had “equity investments of $99 billion and equity investment commitments of $25 billion as of July 26, 2026”. Its commitments table lists $279 billion of supply and capacity, $29 billion of cloud service agreements, $25 billion of data center leases not commenced, $25 billion of equity investments and $8 billion of capital expenditures, a $366 billion total.
A separate table adds $36 billion of AI cloud agreements and $20 billion of data center leases for third parties. Under the AI cloud agreements, the filing says, the clouds buy Nvidia’s infrastructure and Nvidia commits to cloud service agreements, which the AI clouds “can unilaterally stop providing to us and sell to third-party customers at more advantageous rates”.
The guarantees table puts maximum gross exposure at $108.5 billion. That is $3.5 billion of land, power and shell guarantees for AI clouds plus $105 billion of SB Energy guarantees, signed in August 2026 on behalf of an affiliate of OpenAI. They cover leases for about 4.25 gigawatts at a campus in Pike County, Ohio.
As we read the filings, the stake book kept growing through the July quarter while the amount added to it each quarter shrank, from $17.9 billion to $13.1 billion.
As we read the commitments note, more of Nvidia’s exposure now sits in supply contracts and guarantees than in stakes: $279 billion of supply and capacity commitments and up to $108.5 billion of guarantees, against $47.9 billion of private stakes.
As we read Bloomberg’s opening, it does not say whether the slower new agreements Bloomberg Intelligence refers to are sales, investments or partnerships.
NVIDIA Form 10-Q, quarter ended 26 July 2026
“These strategic commitments and guarantees may impact our financial results and are dependent on the performance of our customers and partners.”
Three Implications
FOR NVIDIA’S EARNINGS The 10-Q says unrealized gains, losses and impairments on these stakes are recognized in Other income, net; the stakes carried $9.1 billion of cumulative gross unrealized gains at 26 July 2026.
FOR AI CLOUDS Under the AI cloud agreements, Nvidia commits to cloud service agreements that the clouds can stop providing to it and sell to third-party customers, the 10-Q says.
FOR OPENAI The SB Energy guarantees, capped at $105 billion, support leases for about 4.25 gigawatts on behalf of an OpenAI affiliate, per the 10-Q.
The framework argues: “The $1T capex cycle is being underwritten by a $1T concentrated demand commitment from two counterparties whose combined free cash flow is deeply negative.”
As we read it, Nvidia’s $105 billion SB Energy guarantee cap, given on behalf of an OpenAI affiliate, is one place in a supplier’s own filing where that concentration shows up. The framework predates the guarantee and does not discuss it.
What Is Not Established
We read the opening sentence of the Bloomberg News article, not the Bloomberg Intelligence note. We do not know which agreements Bloomberg Intelligence had in mind; linking its view to the investment figures is our reading. Nvidia’s filings do not name the companies behind the stake totals, and the latest figures are as of 26 July 2026. We did not contact Nvidia or Bloomberg Intelligence.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
Bloomberg Intelligence ranks slower new agreements and China scrutiny as Nvidia’s larger risks. In the filings, Nvidia’s private stakes reached $47.9 billion in July, about 12.6 times a year earlier, while quarterly net additions fell 27% to $13.1 billion and supply commitments rose to $279 billion.
A note on sourcing. Bloomberg Intelligence’s view comes from the opening sentence of a Bloomberg News article of 12 November, the part we could read without a subscription; we did not read the Bloomberg Intelligence note itself. Every Nvidia figure comes from Nvidia’s 10-Q filings for the quarters ended 26 October 2025, 26 April 2026 and 26 July 2026 and its fiscal 2026 10-K on SEC EDGAR, which we read in the relevant notes.
The fourth-quarter figure, the 27% fall and the 12.6 times multiple are our arithmetic. We did not contact Nvidia or Bloomberg Intelligence.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
