One of the most-anticipated days of the second quarter has arrived. Amid a flurry of economic data and President Donald Trump regularly changing his administration's tariff and trade policies, Wall Street's artificial intelligence (AI) darling Nvidia (NVDA 3.23%) is set to report its fiscal 2026 first-quarter operating results after the closing bell today (May 28).
A lot of investors tend to focus on the bread-and-butter headline numbers, which includes Nvidia's revenue and earnings per share (EPS) for the fiscal first quarter, as well as its second-quarter and full-year guidance. These headline figures often provide an extremely quick way for investors to size up how a company performed and/or will perform moving forward.
But if you're just awaiting Nvidia's sales and EPS figures, you'll be missing the big picture. One beneath-the-surface operating metric is what has the potential to make or break Nvidia stock on May 28.
Image source: Getty Images.
Nvidia didn't become Wall Street's leading AI stock by accident
Heading into 2023, Nvidia was a reasonably important tech company with a $360 billion market cap that was prominently known for its graphics processing units (GPUs) used in high-performance gaming and cryptocurrency mining. As of this writing, it's a $3.3 trillion juggernaut that's arguably the most-important tech stock on the planet.
For more than two years, Nvidia's Hopper (H100) GPU and successor Blackwell GPUs have been the backbone of AI-accelerated data centers. Combined, these two chips account for the lion's share of GPUs currently deployed in high-compute data centers.
But Nvidia isn't winning solely because of its market share -- it's excelling due to the law of supply and-demand. Businesses eager to build out their AI-data center infrastructure have found AI-focused GPUs are in short supply. Even with world-leading chip fabricator Taiwan Semiconductor Manufacturing expanding its production capacity at a breakneck pace, demand for Nvidia's chips continue to swamp their supply.
When demand for a good or service outpaces supply, it's expected that the price of said good or service will rise until demand tapers. Nvidia has had little issue raising its prices and charging two to four times more per chip than its external rivals. That's excellent news for the company's sales and profits.
Nvidia's innovation has helped keep it (and its stock) atop the pedestal, as well. CEO Jensen Huang aims to bring a more powerful and energy-efficient chip to market annually through 2027. Following the Hopper and Blackwell are Blackwell Ultra, Vera Rubin, and Vera Rubin Ultra, the latter two of which will be run on a new processor. No external competitors are particularly close to dethroning Nvidia in terms of compute abilities.
Image source: Getty Images.
Everything rests on this operating metric on May 28
Based on what recent history has shown, Nvidia has a better-than-average chance of surpassing Wall Street's consensus sales and EPS forecast for the fiscal first quarter. Nvidia has topped analysts' consensus EPS forecast for nine consecutive quarters.
But there's another metric that tells a more comprehensive story about Nvidia's operating health: its generally accepted accounting principles (GAAP) gross margin.
Gross margin (also called "gross profit margin") is a simple calculation of revenue remaining (expressed as a percentage) after deducting for cost of goods sold. For example, if you brought in $100 in sales and your cost of goods was $20, your gross margin would be 80% (100 – 20 = 80, and (80/100 X 100) = 80%).
Nvidia's GAAP gross margin soared after the Hopper became the hottest thing since sliced bread. Being able to charge more than $40,000 per chip in early 2024 sent its GAAP gross margin to a peak of 78.4% during the fiscal 2025 first quarter, which in turn helped Nvidia's net income skyrocket.
NVDA Gross Profit Margin (Quarterly) data by YCharts.
However, Nvidia's GAAP gross margin has changed course since the comparable quarter one year ago (all quarters based on Nvidia's fiscal year, which ends in late January):
- Q1 2025: 78.4%
- Q2 2025: 75.1%
- Q3 2025: 74.6%
- Q4 2025: 73%
- Q1 2026: 70.6% est. (+/- 50 basis points)
To be objective, a GAAP gross margin of 70.6% would still about 500 basis points higher than where Nvidia's gross margin had been vacillating at the high-end over the previous 10 years before the AI boom. Further, the company is generating considerably more in net sales now than it was even one year ago when it had a 78.4% GAAP gross margin.
Nevertheless, a trend is being set -- and not an encouraging one.
Nvidia's (expected) four-quarter decline in GAAP gross margin looks to be directly related to increasing competition in AI-accelerated data centers. While some of this competition is external, what's more worrisome for Nvidia is that many of its core customers by net sales are developing their own AI-GPUs and AI solutions. Even though these internally developed chips and solutions can't hold water next to the compute potential of Nvidia's hardware, they're considerably cheaper and not backlogged. In other words, it's a recipe for Nvidia to lose out on future data center real estate.
The heart of the problem for Nvidia is that this combination of external and internal competition is steadily (pardon the apropos pun) chipping away at its biggest advantage: its pricing power. Wall Street's AI darling has relied on AI-GPU scarcity to charge an exorbitant premium for its Hopper and Blackwell chips. With this scarcity now diminishing, and expected to wane in future quarters, there's the real possibility of Nvidia's pricing power fading. Chances are that we're already witnessing this in the company's GAAP gross margin decline.
When Nvidia reports its fiscal first quarter results after the bell on May 28, its GAAP gross margin will tell the tale. If its second quarter GAAP gross margin forecast points higher than what's reported for the first quarter, Nvidia stock can breathe new life. But if the forecast calls for a fifth consecutive quarterly decline in GAAP gross margin, Nvidia stock could have further to fall.
