For much of 2026, it’s been a strange time to be Nvidia. Its business kept growing, but investors became harder to convince.
That was, until this week.
Shares of Nvidia (NVDA) jumped 6.8% on Aug. 27 after the artificial-intelligence chipmaker did something it rarely does. Management looked well beyond the next quarter and informed investors of the growth it still expects ahead.
Nvidia anticipates sales to climb about 70% in its next fiscal year ending in January 2028. The unusually long-range estimate helped ease one of the greatest anxieties around the AI trade: that expenditure on data centers and AI infrastructure is approaching a peak, Reuters said.
The reaction was massive. Nvidia was on track to add about $296 billion in market value in a single session, Reuters noted.
And the rally wasn’t limited to Nvidia.
Other AI-related chip firms gained more than $150 billion in value, with Intel, Micron, Broadcom, and U.S.-listed shares of SK Hynix rising between 1.3% and 3.5%. Cloud startups CoreWeave and Nebius, funded by Nvidia, also rose.
Suddenly, it was a simple warning from investors: The AI infrastructure boom may be closer to over.
“AI has reached its inflection point. It’s doing useful work,” Nvidia CEO Jensen Huang said.
Nvidia gives Wall Street numbers it can’t easily ignore
The stock may have been in play on the longer-term view, but Nvidia’s latest quarterly results illustrate why investors took it seriously.
For the fiscal second quarter, revenue was $96.2 billion, up 106% year over year and up 18% sequentially.
Data Center revenue was a whopping $89 billion, up 117% year over year.
Nvidia also reported GAAP operating income of $63.7 billion and net income of $59.7 billion. Adjusted diluted earnings per share were $2.22. Gross margin was unchanged at 75%.
Related: Nvidia customers face 15% AI price shock
For some context, Nvidia had $46.7 billion in total sales the same quarter last year. Thus, in 12 months, it gained about $50 billion in quarterly sales.
The corporation isn’t signaling a slowdown anytime soon, either.
For the fiscal third quarter, Nvidia expects the following:
- Revenue of about $108 billion, plus or minus 2%
- GAAP and adjusted gross margins of roughly 74%
- Operating expenses of approximately $9.2 billion on a GAAP basis
- No Data Center compute revenue from China included in its outlook
Those metrics matter to far more than just Nvidia shareholders.
The world’s biggest tech companies have poured vast sums of money into establishing AI infrastructure, raising worries whether firms such as Microsoft, Alphabet, Amazon and Meta will be able to make enough money from AI to justify their investment.
Nvidia effectively stands at the cash register of that infrastructure boom.
Nvidia keeps selling processors, networking equipment, and increasingly comprehensive artificial intelligence systems if clients keep building.
That makes Nvidia’s 70% growth forecast more than corporate guidance. It is essentially a projection of where the greater AI economy is going.
Nvidia’s next chip is already becoming the center of the story
Vera Rubin, another reason Wall Street went crazy, is Nvidia’s next-gen AI platform.
Rubin is now moving into full production, with systems running at partners including Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure, CoreWeave, and Nebius.
At least 16 brokerages upped their Nvidia price estimates after the results. The analysts’ optimism is being fueled by demand for Rubin.
More Nvidia:
- Nvidia just made a move Wall Street wasn’t ready for
- Nvidia just locked down deal that changes AI race
- Nvidia stock is doing something it hasn’t done in years
The company is also expanding how it seeks clients.
It’s no longer only a few major tech companies driving demand for AI infrastructure, says Huang. The buildout is gradually being joined by frontier AI laboratories, startups, governments, industrial corporations, and enterprises.
And that diversification is important because one of the greatest issues about Nvidia has been concentration.
The economics are tough for competitors to overcome.
Nvidia’s forward price-to-earnings multiple was about 17.9 times, compared with 37.2 for Advanced Micro Devices and 46.2 for Intel, Reuters reported. This reflects how rapidly analyst earnings estimates for Nvidia have risen relative to its share price.
That valuation comparison is eye-popping for a company that merely doubled its quarterly revenue.
Nvidia’s monster forecast still carries a major risk
But there is another aspect to the remarkable success of Nvidia.
Producing enough hardware to fulfill the AI industry’s needs has become a challenge.
Nvidia’s supply chain commitments have reportedly increased to about $279 billion, according to The Wall Street Journal. This more than doubles a prior $119 billion level, as the company seeks to lock in essential components, including high-bandwidth memory.
Memory supply matters most because if vendors cannot deliver enough advanced memory and other components, then Nvidia cannot simply make unlimited AI accelerators.
That creates an unusual situation: Supply, not demand, may be Nvidia’s biggest near-term problem.
Already, Nvidia’s gross-margin projection for the third quarter is down to 74% from 75% in the second quarter. Even if revenue growth is strong, higher memory prices could weigh on profitability.
There are other dangers.
OpenAI and a handful of hyperscalers are building their own proprietary AI chips, which could reduce their dependence on Nvidia over the long term. China is another question mark, with U.S. export limits restricting Nvidia’s ability to sell its most powerful processors there.
Nvidia notably assumes zero China Data Center compute revenue in its third-quarter forecast.
And then there is the biggest concern of all: Will the hundreds of billions of dollars spent on AI infrastructure ever yield enough returns for Nvidia’s customers?
Nvidia’s forecast changes the AI debate
Nvidia’s profits are significant to ordinary investors because the business is now a kind of real-time measure of worldwide AI spending.
So the discussion around Nvidia is evolving.
For a good part of 2026, investors weren’t questioning whether AI was expanding; they wanted to know when that growth would start to taper off.
Nvidia suddenly offered them a whole new explanation.
Revenue has doubled. Data Center sales more than doubled. Rubin is going into production. Management is projecting another big growth year.
The bullishness has become so intense that Raymond James analyst Simon Leopold allegedly envisions a route for Nvidia to eventually surpass $1 trillion in yearly revenue, MarketWatch reported. This may happen around the fiscal year ending January 2029, though current consensus projections are well below that.
This figure should not be interpreted as a forecast on which investors can rely. But its presence alone shows how vastly different the scale of the company is.
Three years ago, the question was whether Nvidia would rule the emerging market for AI accelerators.
Now investors are talking about whether a semiconductor business might one day earn something close to $1 trillion in revenue every year.
That’s the encouraging news for Nvidia stockholders, and the reason why expectations may suddenly be nearly as spectacular as the company’s results.
