Nvidia Q2 Revenue Hits $96B on AI Chip Demand, Concentration Risks Persist
The chipmaker's 106% year-over-year growth confirms surging demand, but five customers account for 70% of receivables and the company carries $108.5 billion in customer guarantees.
Strong demand, stretched exposure
Nvidia reported second-quarter revenue of $96.2 billion, representing 106% year-over-year growth, with $89 billion coming from its data center business. CEO Jensen Huang broke with his usual practice and provided forward guidance, forecasting 70% data center revenue growth for fiscal 2028, which begins in late January 2027.
The results, first reported by Forbes contributor Paulo Carvão, address investor concerns about whether AI infrastructure spending would sustain its pace. Huang told CNBC that demand is "super strong" and accelerating, with CFO Colette Kress characterizing the 70% growth forecast as "supply-constrained."
Why it matters
Nvidia's earnings confirm that enterprise AI adoption is broadening beyond the largest cloud providers, but the company's financial structure reveals significant concentration risk. With one customer representing 16% of revenue and five customers comprising 70% of accounts receivable, Nvidia's exposure to a handful of buyers creates vulnerability. The company has extended payment terms to up to one year for select investment-grade clients, pushing days sales outstanding to 60 days. More concerning, Nvidia carries $108.5 billion in customer guarantees, with $105 billion tied to a single 4.25-gigawatt OpenAI project in Ohio. These financing arrangements effectively make Nvidia a lender to the AI infrastructure buildout it supplies.
Demand accelerates beyond hyperscalers
Hyperscale customers—the major cloud providers supporting AI labs like Anthropic and OpenAI—purchased $48.7 billion worth of products, up 13% from the previous quarter. More notably, the AI clouds, industrial customers, and enterprises segment grew 25% quarter-over-quarter and 138% year-over-year to reach $40.3 billion. The faster growth outside hyperscale customers indicates broader enterprise adoption.
Gross profit margins declined from 75% to an expected 74% in the current quarter, with further erosion to 71-72% anticipated in the fourth quarter. Kress attributed the margin pressure primarily to memory shortages driving higher component costs.
Fastest product ramp in company history
Nvidia's newest Vera Rubin platform reached full production in August 2026 and is already deployed at CoreWeave, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure, and Nebius. The platform is expected to generate 20% of data center revenue in the current quarter, which Kress called "the fastest product ramp in Nvidia's history."
The rapid adoption suggests customers are not delaying purchases to wait for next-generation architecture, reducing the execution risk that typically accompanies major product transitions.
Mobilizing third-party capital
To address concerns about its direct financial exposure, Nvidia announced partnerships with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to mobilize more than $500 billion in third-party capital for customers. The company's 10-Q filing shows $36 billion in "AI-cloud commitments" for helping clients secure data center capacity, land, and power, plus $366 billion in future commitments, mostly from $279 billion for supply and capacity.
Kress emphasized that third-party capital providers will conduct independent underwriting and that Nvidia's compute products are "durable and fungible," addressing analyst concerns about circular financing arrangements.
The details were first reported by Paulo Carvão, a Research Fellow at Harvard, writing for Forbes.
This is an original analysis by the Omega editorial team. Source reporting: AI Watch.
Want systems like this working for your business?
Book a Call