NVIDIA at the Center of the AI Industrial Revolution

The Numbers Are Getting Hard to Ignore

NVIDIA just reported its Q1 FY2027 earnings on May 20, and the headline numbers demand attention: $75 billion in data center revenue, up 92% year-over-year. At a scale where most companies would be cycling decelerating comps, NVIDIA is accelerating.

The full fiscal year just closed with data center revenue of $194 billion — a 68% increase that looks even more staggering when you zoom out. Since ChatGPT emerged in early 2023, NVIDIA’s data center business has grown roughly 13x.

As CFO Colette Kress put it on the call: “We have now scaled our data center business by nearly 13x since the emergence of ChatGPT in fiscal 2023. We look ahead, we expect sequential revenue growth throughout calendar 2026, exceeding what was included in the $500 billion Blackwell and Rubin revenue opportunity we shared last year.”

The Blackwell Story Keeps Getting Better

Blackwell is NVIDIA’s fastest product ramp in company history — an astonishing claim given how fast Hopper ramped before it. In Q1, Blackwell contributed nearly 70% of data center compute revenue, with the transition from Hopper essentially complete. The GB300 NVL72 systems are now in production shipments, and the next-generation Rubin platform — comprising six new chips — is already in fab at TSMC.

CEO Jensen Huang laid out the vision: “Blackwell and Rubin AI factory platforms will be scaling into the $3 to $4 trillion global AI factory build-out through the end of the decade. Customers are building ever-greater scale AI factories. From thousands of Hopper GPUs in tens of megawatt data centers to now hundreds of thousands of Blackwells in 100-megawatt facilities and soon, we will be building millions of Rubin GPU platforms powering multi-gigawatt multisite AI super factories.”

The cadence is now annual. Blackwell → Blackwell Ultra → Rubin → next. Each generation delivers step-function improvements in tokens-per-watt and total cost of inference.

Why Demand Is Not Slowing

Skeptics have been calling the top on AI capex since early 2024. The data keeps proving them wrong. NVIDIA reports that hyperscaler data center capital spending is running at approximately $600 billion annually, having doubled in two years. Meanwhile, AI-native startups — companies that didn’t exist a few years ago — saw their collective revenue go from $2 billion to $20 billion year-over-year.

But the biggest shift is in how AI is being consumed. Inference workloads have overtaken training as the primary driver of GPU utilization. Jensen noted: “One-shot chatbots have evolved into reasoning agentic AI that research, plan, and use tools — driving orders of magnitude jump in compute for both training and inference.”

And the enterprise wave is only beginning. With open-source models maturing, NVIDIA’s RTX Pro servers are being adopted by the likes of Hitachi, Lilly, Hyundai, and Disney. Kress described it as a potential multibillion-dollar product line as enterprises modernize their data centers.

A New Lens on the Business

Starting this quarter, NVIDIA introduced a new reporting framework that better captures the breadth of its opportunity:

• Hyperscale ($38B in Q1): Public cloud and consumer internet companies — still the largest segment, growing 12% QoQ
• ACIE (AI Clouds, Industrial, Enterprise — $37B in Q1): Growing 31% QoQ, with AI cloud revenue more than tripling year-over-year. Sovereign AI revenue alone grew over 80%

This split reveals something important: the “everything else” category is now almost as large as hyperscale and growing faster. The idea that NVIDIA’s fate is tied to 3-4 big customers is outdated.

The Geopolitical Wildcard

No NVIDIA analysis can ignore China export controls. The company took a $4.5 billion charge in Q1 FY2026 related to H20 restrictions, and Q2 saw an additional $4 billion sequential decline in China H20 revenue. The Q3 outlook did not include any H20 shipments.

But NVIDIA is adapting. Management continues to advocate for approved products in China and is diversifying its geographic exposure. The company just announced a $150 billion annual spend commitment in Taiwan and plans for a new campus there that can house four times its current local workforce. The message is clear: regardless of trade policy, NVIDIA intends to build where the supply chain is.

The Macro Environment

Against an economic backdrop where the Fed has cut to 3.64%, the 10-year Treasury sits at 4.5%, and the VIX is a calm 17, the macro climate remains supportive for risk assets. Inflation at 3.78% YoY is stuck above target but not accelerating. The yield curve is no longer inverted — a historically reliable signal that recession fears are fading. Consumer sentiment is low (49.8), but that disconnect between sentiment and actual economic output (GDP at +2.66%) has persisted for over a year without triggering a downturn.

For NVIDIA specifically, the macro bull case is simple: as long as hyperscalers are spending $600B/year on infrastructure and AI-native companies are growing revenue 10x annually, demand for the best compute platform has a long runway.

What It Means

NVIDIA’s story has evolved. It’s no longer about a single chip winning benchmarks. It’s about a full-stack AI computing platform — hardware, networking (Spectrum-X, InfiniBand, NVLink), software (CUDA, TensorRT, Dynamo), and systems engineering — that has become the default foundation for the global AI build-out.

The $500 billion in Blackwell and Rubin revenue visibility (which management says they’re now exceeding) gives NVIDIA something most mega-cap tech companies can’t offer: line-of-sight to 2-3 years of revenue growth at a time when most of the market is guessing about next quarter.

Key Metrics at a Glance

Disclaimer: The above is not investment advice. Please do your own research before making any investment decisions.