Nvidia obliterates Q3 forecasts with $57B quarter, solidifies dominance over AI Supply Chain

You could almost feel the hesitation ripple through Wall Street in the hours leading up to Nvidia’s earnings; investors were openly questioning whether the AI boom had stretched too far. That uncertainty faded the moment the numbers arrived.

Nvidia reported $57 billion in revenue, exceeding expectations of $54.92 billion, and $31.9 billion in earnings, up 65% from $19.3 billion in the previous year. Guidance for the next quarter came in at $65 billion, well above the $61.66 billion projected by analysts, reflecting the strength of ongoing demand and a growing backlog of orders.

This surge is largely driven by sales to hyperscale cloud providers and enterprises building AI infrastructure. CEO Jensen Huang noted that cloud GPUs are sold out, signalling that the company’s chips are the backbone of a rapidly expanding AI ecosystem. The market responded with a more than 4% rise in shares after hours, underscoring how critical these numbers were to investor confidence.

How Nvidia Became the Market’s Core Infrastructure

The clearest sign of Nvidia’s growing dominance comes from its data centre business. In the third quarter, Nvidia reported $51.2 billion in data centre revenue, up 66% year over year. Of that amount, $43 billion came from GPUs powering AI workloads, driven by initial shipments of GB300 chips. Another $8.2 billion came from networking hardware that lets clusters of GPUs operate as a single system.

Blackwell Ultra, the second-generation version of Nvidia’s GPU architecture, became the company’s top-selling product line, showing strong uptake by enterprise and hyperscale clients. S&P Global reports that data centre investment is now the largest contributor to U.S. economic growth, while the “Magnificent 7” tech giants, Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla, now represent 37% of the S&P 500’s total value, illustrating how tightly concentrated this power has become.

This combination, record demand, limited supply, and a buyer base made up of the world’s largest cloud firms, has positioned Nvidia at the centre of the AI infrastructure map.

MORE INSIGHTS ON THIS TOPIC:

• Why SoftBank ditched Nvidia for OpenAI and the AI infrastructure boom
• Nvidia Just Hit $5 Trillion Valuation
• The U.S. government is getting a cut from Nvidia and AMD's AI chip sales to China

A Virtuous Cycle of AI Growth

Huang described the current moment as entering a “virtuous cycle of AI,” pointing to accelerating demand for both training and inference. Nvidia disclosed $500 billion in GPU orders for 2025 and 2026, and hyperscalers, including Microsoft, Meta, Amazon, and Alphabet, collectively raised their capital expenditure forecasts, now expecting to spend more than $380 billion this year on AI infrastructure.

These spending patterns demonstrate why Wedbush analyst Dan Ives called Nvidia’s results a “huge print,” adding that concerns about an AI bubble are “way overstated.” In practical terms, the largest technology companies are investing billions into data centres, servers, and chips, and their spending is already visible in Nvidia’s quarterly performance.

This cycle reinforces itself. As hyperscalers build more AI infrastructure, they require more GPUs. As more GPUs arrive, more AI models and applications are produced, driving further demand. The structure of this cycle naturally favours firms with the capital to keep buying at scale.

Uneven Map of AI Power and Constraints

Demand is heavily concentrated among the largest customers. Nvidia’s $43 billion in compute revenue is driven primarily by its hyperscaler customers. These are the only companies with the financial capacity to deploy AI factories, the scale Nvidia’s CFO says now spans “CSPs, sovereigns, modern builders, enterprises and supercomputing centres.” During Q3 alone, Nvidia announced infrastructure projects totalling an aggregate of five million GPUs.

Smaller organisations cannot match this pace, creating a widening gap in capability and access. Companies that dominate cloud computing are effectively setting the speed and scale of AI innovation, cementing their advantage while smaller players remain constrained by cost and supply.

Not all of Nvidia’s growth is global. The company confirmed it was “disappointed” that it could not ship Blackwell chips to the Chinese market due to U.S. export controls. Although Nvidia received approval to sell its H20 GPU, CFO Colette Kress reported only $50 million in H20 revenue because “sizable purchase orders never materialized.” She attributed the shortfall to geopolitical pressure and growing competition in China. Wedbush’s Ives expects export controls to ease in 2026, but for now, the company’s numbers show it can post record revenue even without participating meaningfully in the world’s second-largest AI market.

Analysts also warn that physical constraints may also emerge. Emarketer’s Jacob Bourne noted that even “explosive growth may not be sustainable” as power, land, and grid capacity threaten to slow data centre expansion. Highlighting that hardware and infrastructure, not chip performance, may set the ceiling for AI expansion. Still, Nvidia’s results indicate demand is outpacing these constraints for now.

A New Tech Oligarchy is Taking Shape

The third-quarter results confirm an ongoing shift in the technology landscape. Nvidia’s surge, driven by record revenue, growing data centre sales, Blackwell Ultra adoption, and $500 billion in forward orders, illustrates how the largest technology companies are consolidating their position in AI infrastructure.

Nvidia has moved from gaming chips to providing the computational backbone for modern AI. As Huang stated, “There’s been a lot of talk about an AI bubble. From our vantage point, we see something very different.” Based on the numbers, that vantage point is hard to argue with. Nvidia’s results show an industry where demand is real, spending is accelerating, and the companies with the largest balance sheets are shaping the next era of computing, one quarter at a time.

Nvidia has become the key supplier powering the AI world, while the hyperscalers control how that power is used. Everyone else, from startups to other countries, relies on them to access AI technology. The gap between these dominant players and the rest is growing, and it’s only set to get bigger.