Nvidia's Moment of Truth: Can the AI Chip King Stay on Top?

.

A Titan Under Pressure

For years, Nvidia has been virtually untouchable. The California-based company built a near-monopoly on the chips that power artificial intelligence — the specialized processors used to train the massive language models behind tools like ChatGPT. But the AI industry is evolving fast, and Nvidia's dominance is no longer guaranteed.

The earnings report due Wednesday (May 21) will be watched closely — not just for the numbers, but for what they signal about the company's future direction. Analysts expect Nvidia to report revenue growth of around 79% for the quarter ending in April 2026, compared to the same period a year ago, according to data from financial research firm LSEG.

From Training to Inference: The Shift That Changes Everything

The AI chip market is undergoing a fundamental shift. For years, the main job of AI chips was training — the computationally intensive process of building AI systems from scratch, which requires enormous processing power. Nvidia dominated this space with an estimated 81% market share, according to market research firm IDC.

But the center of gravity is moving. Increasingly, the demand is for inference — the real-time process of running AI systems, responding to user queries, and executing tasks. Think of it as the difference between building a brain and actually using one. By 2026, roughly 55% of all AI chip workloads are estimated to be inference-related, according to industry data from AI Cloudbase.

This shift matters because the inference market plays to different strengths. It rewards cost efficiency and flexibility — areas where Nvidia faces serious competition.

Who's Challenging the King?

AMD has emerged as the most visible rival. Its data center revenue hit a record $5.8 billion in the first quarter of 2026, up 57% year-over-year, according to company reports. AMD's stock has surged dramatically this year, reflecting growing investor confidence in its AI chip roadmap.

Google (Alphabet) is building an empire within its own walls. The company's custom Tensor Processing Units (TPUs) — chips designed specifically for AI — have secured major contracts worth tens of billions of dollars. Google already runs approximately 20% of its AI inference workloads on its own chips, reducing its dependence on Nvidia.

Amazon is equally aggressive. Its custom Trainium processors have reached an annual revenue run rate of over $20 billion, with purchase commitments of $225 billion, signaling massive future demand. Access to Amazon's Trainium chips is currently fully booked.

Traditional rivals Intel are also repositioning their products to target the cost-sensitive end of the inference market.

Nvidia's Counter-Move

Nvidia is not standing still. In March 2026, the company unveiled a new central processor and AI system built on technology from Groq, an inference-focused startup it acquired. The move signals that Nvidia is serious about competing in the inference space — not just defending its training stronghold.

CEO Jensen Huang has also pointed to the company's upcoming Rubin chip platform, scheduled for later in 2026, as a major growth engine. Nvidia has projected combined revenue of $1 trillion from its Blackwell and Rubin platforms by the end of 2027. The Rubin platform alone is said to deliver up to a tenfold reduction in inference costs compared to current chips.

Nvidia's spending on supply commitments doubled from $50.3 billion to $95.2 billion between its last two fiscal quarters — a sign of aggressive preparation to meet future demand.

The China Problem

One of the biggest wild cards is China. Nvidia once commanded roughly 95% of China's advanced chip market. That market has effectively been cut to zero by a combination of U.S. export restrictions and Beijing's deliberate push toward domestic alternatives like Huawei chips.

The Trump administration approved H200 chip sales to about ten major Chinese companies — including Alibaba, Tencent, ByteDance and JD.com — with each customer allowed to purchase up to 75,000 units. But despite those approvals, not a single chip has been delivered. Beijing has quietly steered Chinese firms away from U.S. chips, directing investment toward domestic alternatives instead.

Jensen Huang joined President Donald Trump's state visit to Beijing earlier this month in what was widely seen as a diplomatic push to unlock the stalled deals. Trump noted that "something could happen," but no concrete breakthrough has been announced. Commerce Secretary Howard Lutnick confirmed that China's central government has, so far, blocked the purchases.

China once accounted for 13% of Nvidia's total revenue, and Huang has estimated the country's AI market alone would be worth $50 billion this year. The continued blockade represents a significant revenue gap.

What Investors Are Watching

Beyond China, several other factors will be scrutinized when Nvidia reports:

Gross margins are expected to come in around 74.5% for the quarter — strong, but analysts warn they could face pressure later in the year due to rising memory costs and the rollout of the new Rubin chip architecture.

Data center capacity is also a bottleneck. Some analysts note that demand for Nvidia chips is so high that customers can't deploy them fast enough — not because the chips aren't available, but because the data centers to house them aren't built yet.

Big Tech spending continues to provide a powerful tailwind. Microsoft, Meta, Google, and Amazon are collectively expected to pour more than $700 billion into AI infrastructure in 2026 — nearly double the spending of just a year ago.

Dominant, But Not Invincible

Nvidia's story remains one of extraordinary success. Its full-year revenue for fiscal 2026 came in at a record $215.9 billion — a 65% increase over the prior year. No other chipmaker is close to those numbers.

But the competitive environment is tightening. Custom chips from Google and Amazon are cheaper for specific inference tasks. AMD is gaining credibility with large cloud customers. And China — once a major market — is actively being closed off by Beijing's industrial policy.

Whether Wednesday's earnings report marks the start of a new chapter or simply confirms Nvidia's continued dominance is a question investors and analysts around the world will be watching closely.

.

Sources:

1. Reuters – Nvidia Q1 2026 Earnings Preview: https://www.reuters.com/world/china/nvidias-outlook-will-be-test-its-strategy-maintain-ai-dominance-2026-05-19/
2. CNBC – U.S. clears H200 chip sales to 10 China firms: https://www.cnbc.com/2026/05/14/us-clears-h200-chip-sales-to-10-china-firms-as-nvidia-ceo-looks-for-breakthrough.html
3. Tom's Hardware – China blocking H200 purchases despite U.S. approval: https://www.tomshardware.com/tech-industry/trump-says-china-is-blocking-h200-purchases
4. Nvidia SEC Filing (Q4 FY2026 Earnings): https://www.sec.gov/Archives/edgar/data/0001045810/000104581026000019/q4fy26pr.htm
5. The Motley Fool – Evidence of Nvidia AI chip dominance under pressure: https://www.fool.com/investing/2026/05/08/the-evidence-is-piling-up-nvidias-ai-chip-dominanc/
6. AI Cloudbase – AI Chip Market Statistics 2026: https://aicloudbase.com/statistics/ai-chip-market-statistics

.