[Summary]
As of May 2026, NVIDIA (NVDA) is not just a semiconductor manufacturer, but one of the world's most important growth stocks at the intersection of AI infrastructure, national AI strategy, data center investment, power infrastructure, and index money flow.
The company's latest official financial results showed sales of $215.9 billion and net income of $120 billion for the full year 2026, which ended January 25, 2026. Fourth-quarter revenue grew to $68.1 billion, and data center revenue grew to $62.3 billion.
The closing price on May 15th was $225.32, and the market capitalization is approximately $5.5 trillion. After hitting a 52-week high of $236.54 on May 14th, it has fallen by 4.42% on May 15th, and in the short term, the stock is in the midst of profit consolidation after the new high and position adjustments ahead of the 2027 first quarter results on May 20th.
The conclusion of this article is clear.
NVIDIA is an “infrastructure provider for the AI era” and at the same time a “super-large growth stock with extreme expectations built into it.”
The bullish factors are overwhelming. However, the stock price is already pricing in huge future growth. From here on, we will look at Blackwell and Rubin's growth, inference demand, Sovereign AI, AI Factory, power constraints, and check the 50-day line and earnings reaction.
First, the conclusion
As of 2026, NVIDIA has the following six faces simultaneously.
| Role | Contents |
|---|---|
| Supplier of AI computing resources | Supplying GPUs and networks for generative AI, inference, and agent AI |
| AI infrastructure company | Integrates not only GPU but also AI Factory, AI Supercluster, network, and software |
| National strategic stocks | Sovereign AI, national AI cloud, the core of data sovereignty |
| Index dominant stocks | Large constituent stocks of S&P 500, NASDAQ-100, and semiconductor ETFs |
| Power Infrastructure Stocks | AI Data Center Power, Cooling, and Grid Investments Inseparable |
| Option supply and demand stocks | Call demand and delta hedges tend to amplify short-term price ranges |
NVIDIA's essence is no longer a "company that sells GPUs."
“A company that supplies the world with capital equipment to power AI”
It is.
Because of this, the market values NVIDIA as central to the AI capital spending cycle, rather than as a regular semiconductor cycle stock.
Evaluation of latest financial results
For the full year 2026, sales were $215.9 billion, operating income was $130.3 billion, and net income was $120.0 billion.
In the fourth quarter alone, the company had sales of $68.1 billion, operating income of $44.2 billion, and net income of $42.9 billion.
Number points
| Item | FY2026 full year | Year-on-year change |
|---|---|---|
| Sales | $215.9 billion | +65% |
| Operating income | $130.3 billion | +60% |
| Net income | $120 billion | +65% |
| Diluted EPS | $4.90 | +67% |
| GAAP gross profit margin | 71.1% | -3.9pt |
| Free CF | $96.5 billion | - |
Data center revenue for the fourth quarter was $62.3 billion, up 75% year-over-year and 22% sequentially. Full-year FY2026 data center revenue was $193.7 billion.
This is the most important point.
NVIDIA's revenue structure has already completely shifted from being centered around gaming GPUs to being centered around data center AI infrastructure.
The next settlement is the biggest event
NVIDIA is scheduled to announce its 2027 first quarter financial results on May 20, 2026.
The company's Q1 FY2027 guidance is for sales of $78 billion, plus or minus 2%. The company explained that this forecast does not include data center compute sales for China.
| Items to check | Reasons to watch |
|---|---|
| Q1 sales | Can we exceed $78 billion guidance |
| Gross profit margin | Can it be maintained at around 75% |
| Blackwell Supply | To what extent will supply constraints, rather than demand, be resolved |
| Demand for inference | Will inference, not just training, become the next growth pillar? |
| Impact on China | Export restrictions and H20-related uncertainties |
| Q2 Guidance | Stock prices are more sensitive to future outlook than past performance |
The May 20th financial results will be an event that will determine the direction not only of NVIDIA alone, but also of the NASDAQ, S&P 500, semiconductor stocks, and AI infrastructure stocks as a whole.
Blackwell and Rubin
Market attention has already shifted from the Hopper generation to Blackwell, Blackwell Ultra, and Rubin.
NVIDIA announced the Vera Rubin platform at GTC in March 2026. Rubin is a concept that integrates GPU, CPU, NVLink, network, DPU, Ethernet switch, etc. and scales AI Factory in rack units.
What is important is that NVIDIA is shifting its focus from competing in the performance of individual chips to the following "system competition."
GPU
↓
NVLinkKey point
↓
DPUKey point
↓
CUDAKey point
↓
AI Factory / AI SuperclusterThis degree of integration is the source of NVIDIA's premium rating.
Even if AMD and proprietary ASICs catch up with GPUs alone, NVIDIA is competing with a full stack that includes GPUs, networks, software, development environments, and operational know-how.
CUDA moat
NVIDIA's biggest moat is CUDA.
CUDA is a software platform for using GPUs for general-purpose calculations and AI processing. NVIDIA's official explanation also states that CUDA is a software layer that allows applications to take advantage of the power of GPUs, and CUDA Toolkit provides compilers, libraries, and development tools.
This is a very difficult point for competitors.
| Competitive factors | NVIDIA's strengths |
|---|---|
| Hardware | Integrates GPU, CPU, DPU, and network |
| Software | CUDA, CUDA-X, AI Enterprise, TensorRT, Triton |
| Developer | Accumulation of researchers, AI companies, cloud, and HPC |
| Operation | Design and optimization of large-scale AI clusters |
| Ecosystem | Extensive collaboration between cloud, servers, data centers, and industrial AI |
Even if competition only improves GPU performance, it will also be necessary to replace the libraries used by AI developers, existing code, model optimization, inference operations, and clustering.
That's why NVIDIA is not just a semiconductor manufacturer.
“A company that holds the development foundation in the AI era”
It is rated as.
Sovereign AI
One of the reasons the market gives NVIDIA a high evaluation is Sovereign AI.
Sovereign AI is a trend in which countries and regions build their own AI computing infrastructure in order to protect their own data, culture, industry, and security.
Oracle and NVIDIA explain that they will deploy the Sovereign AI solution, which allows countries and organizations to operate AI Factory within their own countries and in secure environments.
This means that AI investment is not just a private cloud investment, but a national CAPEX.
Key point
↓
Key point
↓
Key point
↓
GPUKey point
↓
NVIDIAKey pointNVIDIA is both a company selling AI infrastructure and a capital equipment supplier in the race for national AI supremacy.
Power infrastructure issues
GPUs aren't the only next constraint for the AI market.
Power, cooling, grids, transformers, gas turbines, nuclear power, renewable energy, and data center construction themselves are bottlenecks.
In April 2026, the IEA predicted that data center power demand would increase by 17% in 2025, data center power consumption would double by 2030, and power use of AI-specific data centers would triple.
For this reason, the following perspectives are important when considering NVIDIA's growth.
| Constraints | Implications for NVIDIA |
|---|---|
| GPU supply | Blackwell, Rubin shipping capacity |
| HBM/Advanced packages | Memory and CoWoS system supply constraints |
| Electricity | Physical upper limit of AI Factory construction |
| Cooling | High-density rack operating conditions |
| Power grid | Data center connection delay risk |
| Regulations | Local residents, electricity rates, environmental regulations |
NVIDIA itself, with its Vera Rubin DSX AI Factory reference design, points the way to designing an AI Factory that integrates power, cooling, networking, storage, and operations.
In other words, NVIDIA is not only competing on GPUs, but also on power efficiency and AI Factory design itself.
Index domination and passive funds
NVIDIA's market capitalization is approximately $5.5 trillion as of May 15, 2026.
NVIDIA has a very large presence in the S&P 500, NASDAQ-100, semiconductor ETFs, and AI-related ETFs.
This structure facilitates the following self-reinforcing capital flows:
NVDArise
↓
rise
↓
ETFKey point
↓
Key point
↓
Key pointOf course, this is not just an upward trend.
If NVDA collapses, the entire index, ETFs, semiconductor stocks, and AI-related stocks are likely to reverse rotation.
Therefore, NVIDIA is currently not a "stock that is influenced by the index", but
“Stocks that move the index”
It has become.
Option supply and demand
NVIDIA is a huge player not only in the spot stocks but also in the options market.
In particular, demand for call options tends to increase around the time of settlement of accounts, when prices are hitting record highs, and when AI materials are introduced.
At this time, market makers may buy physical stocks to manage risk.
share pricerise
↓
Key point
↓
Key point
↓
share pricerise
↓
Key pointHowever, this also works in reverse.
If expectations are not met after the closing, unwinding of calls or unhedging may amplify the decline.
NVIDIA is a stock that tends to fluctuate in price both when it rises and when it falls.
Technical analysis
After hitting a 52-week high of $236.54 on May 14th, it closed at $225.32 on May 15th.
In the short term, it includes profit taking after a new high and risk adjustment before settlement.
Important line
| Price range/indicators | How to view |
|---|---|
| $236.54 | 52-week high on May 14th |
| $225.32 | Closing price on May 15th |
| Near $220 | Short-term confirmation line of the recent uptrend |
| 50-day moving average line | Important line for institutional investors, CTAs, and momentum funds |
| 200-day moving average | Long-term upward trend bedrock line |
The most important thing for NVIDIA is the 50-day line.
For large momentum stocks like the current one, the 50-day line is more likely to act as a re-inflow line for buying on the dip, rather than just an average on the chart.
However, if the price clearly falls below the 50-day line after settlement, it may lead to a short-term risk-off for the entire AI market.
Valuation
If you simply compare the May 15th closing price of $225.32 and FY2026 GAAP EPS of $4.90, the actual P/E ratio is approximately 46 times.
225.32 / 4.90 / 46According to StockAnalysis data, the PER is in the 48x range and the Forward PER is in the 28x range.
This is a high valuation, but it is based on actual revenues of $215.9 billion in FY2026 sales, $120 billion in net income, and $96.5 billion in free cash flow.
In other words, NVIDIA is not a "bubble stock that is bought only in dreams."
On the other hand, the stock price has already priced in a significant amount of the upcoming growth.
| Conditions to justify evaluation | Points to check |
|---|---|
| Blackwell supply expansion | Will shipments catch up with demand |
| Rubin Roadmap | Is the transition going smoothly after the second half of 2026 |
| Inference demand | Will token generation demand lead to profit growth |
| Continuation of AI CAPEX | Investment stance of hyperscalers |
| Sovereign AI | Continued national and regional cloud investment |
| Overcoming power constraints | Is it possible to overcome the physical constraints of AI Factory construction |
To tolerate a high P/E ratio, growth rates and gross profit margins must continue to be maintained.
Structural risk
The biggest medium- to long-term risk is in-house production of AI computing resources.
Big tech companies are now relying on NVIDIA to power their own AI semiconductors.
| Competition/in-house production | Points to watch |
|---|---|
| Alphabet TPU | Google cloud/in-house AI workloads |
| Amazon Trainium | AI inference/learning demand within AWS |
| Microsoft Maia | In-house AI chip for Azure |
| Meta's original ASIC | Inference cost reduction |
| AMD MI series | Externally sold GPU competition |
| Huawei Ascend | Alternative demand in the Chinese market |
However, for now, NVIDIA's advantage is backed by CUDA, networking, software, rack-scale design, and developer ecosystem.
For competitors to defeat NVIDIA, they will need to replace not only chip performance, but also AI development, inference operations, cluster design, and software optimization.
Short-term risk
The biggest event in the short term is the May 20th financial results.
The risks you should be aware of are as follows.
| Risk | Impact |
|---|---|
| Q1 sales/Q2 guidance expectations not met | Reversal of high PER |
| Gross profit margin decline | Be wary of Blackwell startup costs and mix deterioration |
| Chinese regulations | Impact of H20 and export regulations |
| Supply constraints | HBM, CoWoS, network components |
| Power constraints | Slowing pace of AI Factory construction |
| Option rewind | Price range expansion after settlement |
| Index concentration | NVDA decline spread to entire index |
NVIDIA is one of the strongest stocks, but it also has one of the strongest expected values.
Even if the financial results are strong, a stock may be sold if the guidance and comments do not meet the market's excess expectations.
Second half of 2026 scenario
Main scenario
Blackwell supply expands, Rubin expectations maintain, AI CAPEX and Sovereign AI demand continue to play out.
The conditions are as follows.
| Conditions | Contents |
|---|---|
| Financial results | Q1 FY2027 exceeds $78 billion guidance |
| Guidance | Demand for AI infrastructure continues to show strength in Q2 and beyond |
| Inference | Demand for agent AI and inference will become a new growth pillar |
| Electricity | Absorbing constraints in AI Factory construction through efficiency improvements |
| Supply and demand | ETF, passive, option supply and demand support new high prices |
If these conditions are met, NVIDIA could retest its 52-week high of $236.54 and lead the risk-on trend in the overall AI market.
Risk scenario
After the financial results, expectations are not met, the gross profit margin declines, disappointment with the Q2 guidance, or there is increased vigilance over China, electricity, and supply constraints.
In this case, we will assume an adjustment to the 50-day line, and if the price falls below the 50-day line, we will also assume risk-off for semiconductor stocks as a whole.
However, due to the huge demand for data centers, actual profits in FY2026, Sovereign AI, and the AI Factory concept, at present it is more likely than a collapse of the long-term trend.
“Huge range formation in the high price area”
It is natural to assume that first.
Overall evaluation
As of 2026, NVIDIA is one of the most important stocks in the world at the intersection of the AI revolution, national AI strategy, index dominance, power infrastructure, semiconductor supply chain, and options supply and demand.
The bullish factors are overwhelming.
| Bullish factors | Contents |
|---|---|
| Performance | FY2026 sales $215.9 billion, net income $120 billion |
| Data Center | FY2026 sales $193.7 billion |
| Blackwell | Mainstay of AI infrastructure demand |
| Rubin | Expectations for the next generation AI Factory |
| CUDA | Developer/Software Hori |
| Sovereign AI | National CAPEX |
| Power infrastructure | AI Factory design and energy efficiency are new competitive axes |
On the other hand, the risks are also clear.
| Risk | Contents |
|---|---|
| Valuation | Actual PER approximately 46 times, expectations are factored in significantly |
| In-house production | Hyperscalar original ASIC |
| Regulations | China Export Regulations, Geopolitics |
| Supply constraints | HBM, advanced packaging, power, cooling |
| Reversal of supply and demand | Reverse rotation of options, ETFs, and index links |
In conclusion, NVIDIA
“World capitalism itself in the AI era”
It is no exaggeration to say that Japan is at the center of the world's risk money.
However, when making investment decisions, it is necessary to not only look at the enthusiasm, but also to calmly look at the May 20th financial results, 50-day line, Q2 guidance, power constraints, and in-house production risks.
Final conclusion
The focus after the beginning of the week is
“How much will we exceed the $78 billion guidance in Q1 FY2027 results on May 20th, and how much will we raise our next forecast?”
It is.
On the stock price side, the biggest test will be whether the stock can test the 52-week high of $236.54 again, or whether it can maintain the 50-day line on the downside.
NVIDIA is more than just a GPU company.
“The world's strongest AI infrastructure company that brings together computing resources, software, networks, power efficiency, and national AI infrastructure that power AI”
It is.
However, that strength is already factored into the stock price to a large extent.
The stronger the stock, the more severe it will react to the difference in expectations after earnings. When looking at NVDA in the second half of 2026, it is necessary to simultaneously handle the bullish structure and the volatility that comes from high expectations.
Source
- NVIDIA "Fourth Quarter and Fiscal 2026 Financial Results": https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-Announces-Financial-Results-for-Fourth-Quarter-and-Fiscal-2026/
- NVIDIA “Q1 FY2027 Financial Results Conference Call”: https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-Sets-Conference-Call-for-First-Quarter-Financial-Results/default.aspx
- NVIDIA “Vera Rubin Opens Agentic AI Frontier”: https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-Vera-Rubin-Opens-Agentic-AI-Frontier/default.aspx
- NVIDIA "Vera Rubin DSX AI Factory Reference Design": https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-Releases-Vera-Rubin-DSX-AI-Factory-Reference-Design-and-Omniverse-DSX-Digital-Twin-Blueprint-With-Broad-Industry-Support/default.aspx
- NVIDIA Developer "CUDA Platform": https://developer.nvidia.com/cuda
- NVIDIA/Oracle “Sovereign AI Worldwide”: https://nvidianews.nvidia.com/_gallery/download_pdf/65f879573d6332ddce3719f6/
- IEA “Data center electricity use surged in 2025”: https://www.iea.org/news/data-centre-electricity-use-surged-in-2025-even-with-tightening-bottlenecks-driving-a-scramble-for-solutions
- StockAnalysis “NVIDIA Corporation (NVDA)”: https://stockanalysis.com/stocks/nvda/
- FinanceCharts “NVDA Stock Price History”: https://www.financecharts.com/stocks/NVDA/summary/price