The surge and hidden dangers of "computing power equals revenue": An analysis of NVIDIA's latest financial report

On August 26, Wednesday, NVIDIA announced the strongest quarterly performance in the history of the semiconductor industry. Revenue reached $96.2 billion, with data center revenue at $89 billion, and adjusted earnings per share of $2.22, exceeding expectations by over 6%. The Q3 revenue guidance was set at $108 billion, significantly higher than analyst forecasts. Subsequently, the stock price surged about 4% in after-hours trading. This report will explain what these numbers actually mean, why gross margin is the core issue of focus for investors, and what to watch for next.

Key Data: FY2027 Q2 revenue of $96.22 billion, a year-on-year increase of 106% · Data center revenue of $89.02 billion, a year-on-year increase of 117% · Adjusted EPS of $2.22, higher than the consensus expectation of $2.09 · Q3 revenue guidance of $108 billion, above the consensus expectation of about $104 billion · Q2 gross margin of 75%, Q3 guidance of 74%, and Q4 expected to further decline to 71%-72% · Supply commitments soared from $119 billion to $279 billion · After-hours stock price rose about 4% to $218 · $26 billion returned to shareholders this quarter

Section 1 --- Actual Numbers

Let’s clarify in simple terms what NVIDIA actually reported.

For the quarter ending July 26, 2026, revenue was $96.22 billion, a 106% increase year-on-year, and an 18% increase from Q1's $81.61 billion, exceeding the analyst consensus expectation of about $92.1 billion by 4.5%. The $96 billion quarterly revenue is higher than the annual GDP of many countries and surpasses NVIDIA's total revenue for the entire fiscal year 2023.

Data center business revenue was $89.02 billion, a year-on-year increase of 117%, and an 18% quarter-on-quarter increase. The data center now accounts for about 93% of NVIDIA's total revenue. The combined revenue from gaming, automotive, professional visualization, and edge computing was only $7.2 billion, which is almost negligible compared to the $89 billion data center business.

Adjusted earnings per share were $2.22, exceeding the consensus expectation of $2.09 by about 6.2%. The GAAP earnings per share were $2.46, which included an $8 billion gain from equity investments (positions include Intel and SpaceX). GAAP net profit reached $59.7 billion, a year-on-year increase of 126%. Gross margin remained at 75% under both GAAP and non-GAAP standards, unchanged from the previous quarter and up 250 basis points from 72.5% a year ago.

The Q3 revenue guidance is $108 billion (plus or minus 2%), while the previous analyst consensus was about $104 billion, exceeding by about $4 billion—this forward-looking number is what the market is most concerned about.

NVIDIA founder and CEO Jensen Huang stated in the press release: "AI has reached a turning point. It is doing valuable work. Every token it produces is productive and profitable. Now, computing power is revenue."

Educational Note: NVIDIA's fiscal year precedes the calendar year. When the company reports "FY2027 Q2" results in August 2026, it refers to the three months ending July 26, 2026. This timing convention is worth remembering: financial media typically refer to it as "August 2026 earnings," while NVIDIA labels it as fiscal year 2027; both terms refer to the same set of numbers.

Section 2 --- Where the Growth Comes From

The NVIDIA story in 2026 ultimately revolves around a single driving force: all major tech companies worldwide are racing to build AI infrastructure, and NVIDIA produces the indispensable chips for them.

The four major hyperscale cloud providers—Alphabet, Amazon, Meta, and Microsoft—had a combined capital expenditure of $166 billion for the quarter ending June 2026, a year-on-year increase of 87% and a quarter-on-quarter increase of 27%. Over the ten quarters since early 2024, their combined capital expenditures have grown by 272%. CFO Colette Kress disclosed that the capital expenditures of the top five hyperscale cloud providers are expected to grow from about $800 billion in 2026 to about $1.3 trillion in 2027. Much of this spending is flowing into NVIDIA's data center business.

Within the data center, hyperscale cloud providers contributed $48.7 billion in revenue, a year-on-year increase of 102%. But even more remarkable growth came from other directions—AI cloud and industrial and enterprise customers (ACIE) contributed $40.3 billion, a year-on-year increase of 138% and a quarter-on-quarter increase of 25%. The customer base is genuinely diversifying: AWS plans to deploy an additional 2 million NVIDIA GPUs, SpaceX stated it will build its AI infrastructure entirely based on NVIDIA, and government-led sovereign AI projects have also become a significant source of revenue.

One of the most noteworthy numbers in the entire earnings report is not a revenue item. Supply commitments surged from $119 billion at the end of Q1 to $279 billion at the end of Q2—more than doubling in a single quarter—primarily related to memory procurement needed for the Vera Rubin platform. This locks in a substantial amount of future revenue that has yet to be shipped and is currently the clearest evidence that the demand NVIDIA sees is not a forecast but a committed order book.

The next-generation platform Vera Rubin began mass production as scheduled in Q3, expected to account for about 20% of Q3 data center revenue. The CFO guided that revenue growth for fiscal year 2028 is expected to be around 70%, clearly stating that this is a result of supply constraints. Jensen Huang said: "Although demand is far beyond 70%, our supply allows us to be confident in achieving 70%."

Section 3 --- Gross Margin: Why This is the Most Watched Number

In after-hours trading, when the complete picture—revenue exceeding expectations, EPS exceeding expectations, $108 billion guidance, $279 billion supply commitments—was presented to investors, the stock price immediately rose about 4% to around $218. However, amid this positive reaction, the most debated issue was not the top-line revenue but the trend of gross margin.

NVIDIA guided Q3 gross margin to be 74%, down 1 percentage point from the recently achieved 75%. At a revenue scale of $108 billion, a 1 percentage point drop in gross margin means a decrease of over $1 billion in quarterly gross profit. The CFO further guided Q4 gross margin to be between 71% and 72%.

The underlying reason is memory costs. Management clearly stated that the increase in memory costs has exceeded previous expectations and is expected to be "even higher" next year. In the $279 billion supply commitments, most are related to memory procurement—the high-bandwidth memory (HBM) prices embedded in every NVIDIA GPU are still rising. Earlier this month, NVIDIA notified customers that it would raise AI server prices by over 15% to cope with soaring memory costs, but the cost increase has outpaced the price pass-through to customers, leading to significant pressure on gross margins.

This directly echoes what we recorded in the memory supercycle report and the SK Hynix report. SK Hynix achieved a 72% operating profit margin in Q1 2026, partly because NVIDIA and its customers had no choice but to purchase HBM at ever-increasing prices. The cost pressures on one company are precisely the revenue tailwinds for another.

Why is gross margin so important even when revenue continues to grow robustly? A simple example illustrates this. If NVIDIA's revenue increases from $96 billion to $108 billion, but gross margin drops from 75% to 74%, gross profit increases from about $72 billion to about $79.9 billion—still growing, but investors modeling $108 billion revenue at a 75% gross margin originally expected about $81 billion in gross profit. A gap of over $1 billion each quarter, compounded over the next several quarters, is not negligible given NVIDIA's scale, which explains why analysts asked numerous questions about gross margin during the conference call, even though the revenue guidance exceeded expectations by about $4 billion.

Educational Note: Gross margin is the ratio expressed as a percentage of revenue after subtracting the direct costs of products. For every $1 of revenue NVIDIA collected in Q2, it retained $0.75 after paying for components, manufacturing, and assembly costs. When this ratio drops to 0.74 in Q3 and possibly to 0.71 to 0.72 in Q4, even a few cents difference per dollar means billions of dollars given NVIDIA's revenue scale. A hardware company with a 75% gross margin is extremely rare—Apple achieves 49%, and most chip companies are in the 50% range. The most pressing question for investors now is: where is the bottom of this number?

Section 4 --- What the Macro Numbers Tell You

Stepping back from the quarterly numbers, several structural facts about NVIDIA's business become clear.

The business is highly concentrated in AI infrastructure. Data centers account for about 93% of total revenue, and NVIDIA has effectively become a single-product company in terms of performance drivers. If spending on AI infrastructure slows for any reason, the revenue impact will be immediate and significant.

Customer concentration is real but improving. Hyperscale cloud providers remain the primary buyers, contributing $48.7 billion; however, ACIE is catching up with $40.3 billion and a faster growth rate, which is a positive signal of structural improvement. A more diversified customer base will reduce the impact of a single buyer's decisions on NVIDIA's performance.

Growth is constrained by supply, not demand. Jensen Huang's wording is clear: demand "far exceeds" the 70% growth guidance under conditions without supply constraints. NVIDIA has more orders than capacity, and any supply chain disruption will immediately affect revenue realization.

The $279 billion commitment is the most underestimated number. The supply commitments doubled from $119 billion to $279 billion in a single quarter, primarily related to memory procurement, conveying confidence in future demand more directly than any verbal guidance. This is real capital, committed to real supply partners.

Shareholder returns are substantial. NVIDIA returned $26 billion to shareholders this quarter through buybacks and dividends, with $99 billion in buyback capacity remaining, while also issuing $25 billion in senior unsecured notes. With a quarterly free cash flow of $21.1 billion, even facing gross margin compression, NVIDIA's free cash flow in one quarter exceeds the annual levels of most S&P 500 companies.

Section 5 --- The Most Important Statement from Jensen Huang

"AI has reached a turning point. It is doing valuable work. Every token it produces is productive and profitable. Now, computing power is revenue."

The phrase "computing power is revenue" is new and deliberate. This is the first clear statement that companies are not only investing in AI infrastructure but that this infrastructure is already generating valuable, quantifiable economic returns. This is precisely the shift the entire investment community has been waiting for—a leap from AI being an investment logic to AI being a profitability logic.

If Jensen Huang's judgment is accurate, this is the most compelling structural argument for sustained demand. If it is premature—if the time required for the AI infrastructure built between 2024 and 2026 to generate economic returns is longer than expected—the capital expenditure cycle driving NVIDIA's extraordinary growth may reach a turning point. The answer to this question will determine whether the current wave of AI infrastructure construction represents a lasting structural shift or a temporary investment cycle that will ultimately moderate.

Section 6 --- How This Relates to Everything We Reported Previously

NVIDIA's FY2027 Q2 results are deeply connected to almost every major theme in this series of reports.

The $166 billion in quarterly capital expenditures from hyperscale cloud providers is the same force driving AI-related stocks higher, fueling the memory supercycle, leading to SK Hynix's record Q1 performance, while simultaneously compressing NVIDIA's gross margins through rising HBM prices.

The absence of the Chinese market directly echoes the CXMT report—export controls have reduced China's data center computing business to less than 1% of NVIDIA's data center revenue, but have also created market space for CXMT to fill domestic AI infrastructure demand.

The $279 billion memory supply commitment is directly linked to SK Hynix's IPO report and the DRAM ETF report—NVIDIA's memory procurement at this scale is one of the main demand drivers that gives SK Hynix, Samsung, and Micron such strong pricing power.

The trend of gross margins aligns with the logic we discussed in the Federal Reserve and yield reports. In an environment with interest rates between 3.5% and 3.75%, investors need high and stable profit margins to support premium valuations. When margins compress—even if just from 75% to 74%—the valuation math changes accordingly.

Section 7 --- Trends to Watch Next

The Q3 earnings report in late November 2026. The most critical indicator is not revenue, but whether gross margin can stabilize at 74% or continue to decline towards the guidance of 71% to 72%. The trajectory of gross margins will determine whether NVIDIA's current valuation level is justified.

Vera Rubin capacity ramp-up. Vera Rubin is expected to account for about 20% of Q3 data center revenue, and a faster-than-expected ramp-up in Q4 through fiscal year 2028 could offset gross margin pressures; delays would exacerbate the situation.

Memory cost trends. HBM pricing from SK Hynix, Samsung, and Micron directly affects NVIDIA's product costs. Any softening in HBM prices would directly benefit NVIDIA's gross margins—while negatively impacting memory companies currently benefiting from high pricing.

Chinese policy direction. Any changes in U.S. export controls on AI chips to China will have immediate substantive effects. The current guidance assumes that this market contributes zero.

AI monetization data. Jensen Huang said, "computing power is revenue," and the investment community will closely track evidence of AI infrastructure generating quantifiable economic returns.

The expectation of $1.3 trillion in capital expenditures from hyperscale cloud providers in 2027. CFO Kress disclosed that capital expenditures from the top five hyperscale cloud providers are expected to grow from about $800 billion in 2026 to about $1.3 trillion in 2027. If this number holds true, NVIDIA's growth trajectory constrained by supply will continue into fiscal year 2028 and beyond.

Conclusion

NVIDIA's FY2027 Q2 results are extremely strong by any objective standard: revenue more than doubled, data center growth of 117%, guidance exceeding expectations by about $4 billion, and supply commitments soaring from $119 billion to $279 billion, with the stock price rising about 4% in after-hours trading.

However, amid this positive market reaction, investors are most focused on the trajectory of gross margins over the next two quarters: from 75% in Q2, to 74% in Q3 guidance, and then to an expected 71% to 72% in Q4. This stepwise decline reflects the management's clear indication of continuing "upward" pressure from memory costs.

NVIDIA remains the most important single company in the AI infrastructure story. Its revenue scale, software moat built through CUDA, platform leadership from Blackwell to Vera Rubin, and the $279 billion supply commitments representing locked-in future demand are indisputable.

What the current market is pricing is the trend of gross margins—and whether the strong revenue growth can sufficiently counterbalance the compression of profitability per dollar of revenue.

The answer will be revealed in late November.

Data as of August 27, 2026. Source: NVIDIA official press release and SEC Form 8-K (August 26, 2026); NVIDIA official press release via GlobeNewswire; AlphaStreet Q2 FY27 earnings report; CNBC earnings real-time updates; 24/7 Wall St.; Tom's Hardware; BigGo Finance; FinanceFeeds; Benzinga; Kiplinger; RexShares; Robinhood market data; CNN Markets; NextPlatform; Shacknews.

Disclaimer: This article is for market information sharing and investor education only and does not constitute any investment, trading, financial, legal, or other professional advice, nor does it constitute a recommendation, offer, or solicitation for any securities, financial products, or investment strategies. The views, data, and forecasts in this article are based on publicly available information and the author's judgment, which may contain errors, omissions, or deviations, and relevant data and views may be adjusted as market and company information changes. Past performance does not represent future results, and investors should independently assess their own situation and bear the associated risks. Some content of this article was provided or co-authored by external authors and does not represent BIT's official investment views, research conclusions, or endorsement of related views. BIT does not guarantee the accuracy, completeness, or timeliness of the views of external authors. Information regarding NVIDIA (NVIDIA) and other securities is for general market discussion only. Investors should verify relevant information independently before making any investment decisions and make independent judgments based on their own risk tolerance.