
Tech • AI • Robotics
Nvidia has delivered extraordinary growth in the AI boom, turning soaring chip demand into massive revenue, stock market dominance and a wider financing role across the industry.
Since the public release of ChatGPT, Nvidia shares have risen about 14-fold, a surge supported by business results rather than speculation alone. Quarterly revenue has climbed so fast that, on an annual basis, sales roughly doubled in each of the past three years. The company is now approaching $100 billion in quarterly revenue, a dramatic leap from its earlier identity as a graphics-card maker focused on gamers and PC enthusiasts.
Nvidia has become so large that its performance now materially shapes broader US equity returns. The company accounts for roughly 8% of the S&P 500, and since the start of 2023 it has generated nearly 15 cents of every dollar returned by the index. That means investors with broad exposure to US stocks are now heavily exposed to the fortunes of a single chipmaker.
A few years ago, Nvidia was estimated to supply nearly 80% of AI chips sold in the market. That share has fallen to around 60% as more demand is met by custom silicon designed by software companies for their own internal systems. By the end of the decade, the market is expected to move toward a 50-50 split, with about half of AI chips coming from Nvidia and half from in-house designs.
Beyond selling chips, Nvidia is increasingly helping enable the infrastructure needed to use them. One prominent example involves OpenAI, which is developing an 8-gigawatt data center in Ohio, a scale comparable to the output of a couple of midsize nuclear reactors. To help make the project financeable, Nvidia is guaranteeing the land and power lease for 20 years, making lenders more willing to back the buildout.
The logic behind such arrangements is that Nvidia appears more bullish than outside capital markets about how much AI infrastructure will be needed. Rather than wait for investors to fund everything at the desired scale, the company is using its own balance sheet to accelerate deployment. That amounts to a double bet: on continued demand for its chips and on the broader economy’s appetite for AI capacity.
These deals raise concerns because Nvidia sometimes invests in companies that later buy its chips. The company resists the label of circular financing, but the overlap is hard to ignore when an equity-backed customer turns into a major purchaser. Still, there is an important distinction: Nvidia is generally not lending customers money to buy chips directly, but instead offering guarantees tied to future supply, pricing, or infrastructure commitments.
Nvidia’s ability to support such commitments rests on exceptional financial strength. It holds about $100 billion in cash and liquid investments and is expected to generate roughly $200 billion in free cash flow this year. Its largest customers are also major, cash-flow-positive hyperscalers, which provides more resilience than the debt-fueled structures often associated with past tech bubbles.
The company’s central role in the AI boom means any serious reversal could reverberate well beyond the semiconductor sector. If AI demand falls short, Nvidia could be hit both by weaker chip orders and by pressure from guarantees tied to projects or counterparties under strain. Analysts see that as more likely to amplify an economic downturn than to trigger one outright, though the scale of the company now makes the risk macroeconomically significant.
Nvidia has evolved from a chip supplier into a central financial and industrial force behind the AI buildout. Its remarkable performance reflects real demand and enormous cash generation, but its growing reach also means that any future slowdown would carry consequences far beyond one company.
Explain this