NVIDIA financial forecast | 2027 fiscal year Q1 passing line and 3 stock price scenarios after announcement

[Summary]

NVIDIA (NVDA) is scheduled to announce its financial results for the first quarter of fiscal year 2027 after the market close on May 20, 2026, US time. The event will take place early in the morning on May 21, 2026, Japan time.

This year's financial results are not just for one company. This is a market event that will have ripple effects on AI semiconductors, NASDAQ, S&P500, data center investment, and power infrastructure related stocks.

The focus is less on the Q1 performance itself and more on the next Q2 guidance, Blackwell shipments, the transition to Rubin, and the sustainability of AI infrastructure investments through 2027.

In conclusion, there are many bullish factors for NVIDIA. However, the stock price has already factored in high expectations, and there is a possibility that the company will run out of material even with normal good results.

Financial results announcement schedule

In an official announcement, NVIDIA will hold a conference call on May 20, 2026 at 2:00 pm PT and 5:00 pm ET to discuss its first quarter fiscal 2027 results.

The period covered is the quarter ended April 26, 2026.

Additionally, the company's financial results and CFO comments are scheduled to be released around 1:20 pm PT.

In Japan time, it can be viewed as follows.

The latest financial results will be announced after the US market closes, so Japanese individual investors will not see them until the morning of May 21, 2026.

Near-term assumptions: Q4 and company guidance

NVIDIA's previous financial results, the fourth quarter of fiscal year 2026, were very strong.

Officially, Q4 sales were $68.1 billion, up 20% from the previous quarter and 73% from the same period last year. Data center sales were $62.3 billion, an increase of 75% from the same period last year.

Full-year sales for fiscal 2026 were $215.9 billion, an increase of 65% from the previous year.

The company's guidance for Q1 FY2027 is as follows.

Importantly, the company issued a sales guidance of $78 billion without expecting data center compute sales to China.

In other words, while the market is looking at China risks, it is also looking at whether growth in demand for AI infrastructure will continue in the US, Europe, the Middle East, Japan, and other countries.

Wall Street Passing Line

This year's financial results are difficult to evaluate if they are just slightly better than expected.

NVIDIA is already seen as ``a good financial result'', so the market's focus has shifted to the whisper number, or how much it can beat the official consensus.

The main market forecasts are as follows.

S&P Global Market Intelligence's Visible Alpha forecast calls for Q1 sales of $78.5 billion and data center sales of around $72.8 billion.

Meanwhile, market forecasts compiled by The Motley Fool based on Yahoo Finance data call for Q1 sales of $79.17 billion, adjusted EPS of $1.78, and Q2 sales of $87.06 billion.

Therefore, the actual passing line this time can be summarized as follows.

Q1net sales：790hundred million dollarsKey point
Q2870
75%
Blackwell：demandKey point
Rubin：2027yearKey point

Market trends: Is the demand for AI real?

The biggest factor behind NVIDIA's financial results is capital investment for hyperscalers.

Cloud giants like Microsoft, Alphabet, Meta, and Amazon are increasingly investing in AI data centers.

The Motley Fool estimates that the 2026 CAPEX plans of the four major companies will total $725 billion, a significant increase from last year's $410 billion.

At the heart of this investment are GPUs, networking, servers, cooling, and power equipment.

NVIDIA's sales are almost directly linked to this AI infrastructure investment.

Supply constraints continue

While demand is strong, there are constraints on the supply side.

TSMC's advanced packaging, HBM, server assembly, power, cooling, network equipment, etc. are likely to become bottlenecks.

In NVIDIA's financial results, it is not only important to consider whether there is demand, but also how much it can be shipped.

The focus is on risks that cannot be created, rather than risks that cannot be sold.

China Risk and Sovereign AI

Export restrictions to China continue to be a risk for NVIDIA.

However, NVIDIA does not factor in data center compute sales for China in its Q1 guidance.

On the other hand, demand for Sovereign AI is expanding in the Middle East, Europe, Japan, and other countries.

The trend of state-led AI infrastructure development is a new source of demand for NVIDIA.

Three potential risks to be wary of before closing

The latest financial results cannot be judged simply by whether Q1 sales exceed market expectations.

Market expectations are already extremely high, with some predicting sales of over $80 billion.

Therefore, even if the financial results are good, if the next structural risk becomes apparent, it may be perceived as ``exhausted materials'' or ``doubts about future growth.''

1. Rubin migration delay risk

An important part of the roadmap from late 2026 to 2027 is the next-generation platform Rubin.

However, some research companies have pointed out the risk of supply chain adjustments regarding Rubin's launch.

TrendForce cites challenges such as HBM4 certification, network migration from ConnectX-8 to ConnectX-9, power management to address increased power consumption, and optimization of more advanced liquid cooling systems.

The company has lowered its outlook for the Rubin ratio from 29% to 22% and raised the Blackwell ratio from 61% to 71% for NVIDIA's high-performance GPU shipment composition in 2026.

This doesn't mean NVIDIA's growth will stop.

In fact, demand for Blackwell will be strong in the short term, and it is likely that Blackwell will fill in the gaps for Rubin.

However, for investors,

Will the generational change from Blackwell to Rubin proceed smoothly?

will be the evaluation axis in 2027.

Regarding the reported delay of HBM4, there is also a report that NVIDIA says that "planning is on track," so it should not be treated as an officially confirmed delay by the company at this time.

What will be important is how clearly comments will be made on Rubin, HBM4, liquid cooling, and network migration during the conference call.

2. Data center capacity

Even if NVIDIA increases the production of GPUs, it does not necessarily mean that customers will be able to operate them immediately.

AI data centers require server racks, power, cooling, networks, buildings, and power grids.

Capital investment for hyperscalers is increasing, but actual construction and power connections take time.

Therefore, the following bottlenecks may occur on the customer side.

• Data center construction delay
• Lack of electricity/grid
• Delay in liquid cooling support
• Installation/verification period per rack
• Network equipment supply constraints

The risk for NVIDIA is not that demand will disappear, but that customer acceptance capacity will limit the pace of shipments in the near term.

In this financial results, attention will be focused not only on the strength of demand for Blackwell, but also on how steady the actual pace of shipments and operations are.

3. Gross profit margin peaks out

NVIDIA's strength lies in its high gross profit margin in the mid-70% range.

Q1 company guidance calls for non-GAAP gross profit margin of 75.0%, plus or minus 0.5 points.

On the other hand, in full-scale mass production of Blackwell, costs such as advanced packaging, HBM, liquid cooling, and rack scale design tend to increase.

Additionally, during initial mass production, yield rates and supply chain adjustments may affect profit margins.

The points that the market will look at are as follows.

NVIDIA is a company with very strong sales growth, but its stock price is also based on high profit margins.

Therefore, when the gross profit margin starts to decline, the stock price reaction tends to be slow even if the growth rate is high.

Chinese market: H20/H200 and structural risks of domestic substitution

The Chinese market is the most difficult region for NVIDIA to navigate.

Sales of cutting-edge AI GPUs to China have been restricted due to US export regulations.

NVIDIA has been introducing regulation-compliant chips to China, but uncertainty surrounding H20 and H200 continues.

While there are reports of a new framework for H200 exports between the US and China in 2026, it has also been pointed out that China may prioritize domestic chip development and curb purchases of NVIDIA products.

Some reports say that China has indicated that it will not approve the purchase of NVIDIA H200.

Furthermore, the longer the restrictions last, the more incentives Chinese customers will have to move to AI chips from domestic manufacturers such as Huawei.

This is a medium- to long-term market share risk rather than a short-term sales risk.

However, NVIDIA is trying to offset the China risk with Sovereign AI.

Sovereign AI is a trend in which governments and state-owned enterprises develop AI infrastructure tailored to their own language, industry, administration, defense, medical care, etc.

Expanding state-led AI investment in the Middle East, Europe, Japan, and other Asian countries could offset some of the uncertainty facing China.

In this conference call, it will be important to see how much Sovereign AI contributes to sales and how specifically it will be talked about as a future growth driver.

Shift to inference market and competitive risks

In the AI market, the center of gravity is shifting from learning to inference.

Learning is the calculation required to create huge models, and is an area in which NVIDIA is strongest.

On the other hand, inference is the process of actually moving the completed AI model. Everyday uses of AI, such as answering user questions, running AI agents, and generating images and videos, are inferences.

Key point
↓
Key point

In the future, the demand for inference will increase as the use of AI expands.

However, inference tends to be highly competitive due to the emphasis on cost, power, and low latency.

1. ASIC in-house production for major customers

Hyperscalers such as Microsoft, Google, Amazon, and Meta are huge customers of NVIDIA, and potential competitors.

In-house AI chips like the Google TPU, Amazon Trainium, and Microsoft Maia aim to lower costs and power consumption by optimizing for specific workloads.

Although it will not replace all GPU demand, ASICs may absorb some of NVIDIA's demand for inference applications that run large numbers of fixed models on the company's cloud.

2. AMD's second choice

AMD is the most viable general purpose GPU alternative to NVIDIA.

MI300/MI325/MI350 series Instinct GPUs are highly evaluated in terms of memory capacity and cost, and have become an option for hyperscalers to diversify procurement.

Although it is not enough to immediately replace NVIDIA, AMD's presence is increasing in terms of price negotiation power and supply distribution.

3. Inference specialized startup

Inference-specific players like Groq are also attracting attention.

It is a company aiming at the market with a different architecture from NVIDIA, armed with low latency, low cost, and optimization for specific inference processing.

However, inference-specific chips tend to have issues with software, developer ecosystem, cloud adoption, and supply capacity.

NVIDIA is defending against this by selling systems that combine Grace CPU, NVLink, InfiniBand/Ethernet, NIM, AI Enterprise, and inference optimization software, rather than just GPUs.

What investors should be looking at is whether NVIDIA can continue to hold the ``standard for the entire AI factory'' in the inference market, rather than ``competing on individual chips.''

NVIDIA's business model: Why can it maintain high profit margins?

The reason NVIDIA ranks among the top in the world in terms of market capitalization and is able to maintain a gross profit margin in the 70% range is not simply because it sells high-performance GPUs.

The essence of the company is that it is a platform company that controls the infrastructure ecosystem that supports AI development, learning, inference, operation, and even robotics.

From an investor's perspective, NVIDIA's strengths can be broken down into the following four areas.

1. Software lock-in with CUDA

NVIDIA's biggest barrier to entry is not just the GPU itself.

It is a parallel computing platform called CUDA that was introduced in 2006.

AI researchers, engineers, and cloud companies around the world have been creating AI models and high-speed calculation applications based on CUDA for many years.

Therefore, even if AMD or Intel introduces cheap AI chips, customers will still have to incur software migration costs.

CUDAKey point
↓
NVIDIA GPUKey point
↓
Key point
↓
Key point

This structure supports NVIDIA's pricing power.

2. Vertical integration of fabless and AI Factory

NVIDIA is a fabless company that does not have its own factories, and mainly outsources the manufacturing of cutting-edge semiconductors to TSMC and other companies.

This allows the company to focus its management resources on design, software, networks, and system integration without taking on huge factory investment risks.

On the other hand, NVIDIA is currently not just a company that sells individual chips.

In the Blackwell generation, there is an increasing emphasis on proposing an AI data center platform that combines GPUs, CPUs, networks, memory, cooling, racks, and software, the so-called AI Factory.

As a result, the transaction size per customer will be larger than just semiconductor sales, and the unit sales price to hyperscalers such as Microsoft, Amazon, Google, and Meta will likely rise.

3. Expansion to AI software and subscriptions

In addition to selling off hardware, NVIDIA is also expanding its software infrastructure, including AI Enterprise and NIM.

NVIDIA AI Enterprise is a commercial software suite that allows enterprises to run generative AI and AI agents in production environments.

NIM is an inference microservice that makes it easier to deploy AI models in corporate environments, and has the role of reducing the operational burden of AI implementation.

The further this direction progresses, the closer NVIDIA will move toward a model where it does not just sell the GPU once and that's it, but instead continues to earn revenue from software, maintenance, optimization, and inference operations even after installation.

Importantly for investors, high-margin software revenues have the potential to moderate hardware economic cycles in the future.

4. Deployment to physical AI and robotics

NVIDIA is focusing on physical AI as the next market after generative AI.

This is not on-screen AI like chatbots, but AI that operates in the real world, such as in factories, logistics, autonomous driving, and robots.

Omniverse is used as an industrial digital twin and simulation platform, and Cosmos is positioned as a world model platform that supports physical AI development such as robotics and autonomous driving.

Before operating a real factory or robot, a large number of simulations are performed in virtual space, and the intelligence learned there is transferred to the real system.

If this trend gets into full swing, demand for NVIDIA will expand not only to AI data centers but also to manufacturing, logistics, automobiles, defense, medical equipment, and more.

Flywheel structure

NVIDIA's strength can be explained by the following cycle.

Key point
↓
NVIDIA GPUKey point
↓
Key point
↓
Blackwell、Rubin、CUDA、AI Enterprise、OmniverseKey point
↓
Key point

As long as this flywheel continues, NVIDIA will likely be evaluated as not just a semiconductor stock, but a company that holds the infrastructure standards for the AI era.

However, the stronger the lock-in of a company, the higher the market expectations.

That's why in these financial results, we need to look beyond just sales and see if the entire ecosystem, including CUDA, AI Enterprise, Blackwell, Rubin, and physical AI, is expanding.

Are AMD and its own AI chips a threat?

In conclusion, AMD and Big Tech's in-house AI chips are a clear threat to NVIDIA.

However, at this point in time, rather than NVIDIA's stronghold being destroyed all at once, it is more realistic to see the company as the second or third-largest player in the rapidly expanding AI accelerator market.

According to market estimates, NVIDIA will maintain a revenue share of around 75-80% in the AI ​​accelerator market even in 2026.

On the other hand, the nature of competition is changing.

The main battleground so far has been training for creating large-scale AI models.

However, in the future, the interference, or inference, that drives completed AI on a daily basis will become more important.

Specialized chips like AMD, Google TPU, AWS Trainium, and Microsoft Maia are gaining traction in this inference market.

1. AMD: The only serious general-purpose GPU alternative

AMD is one of the few companies that provides general-purpose AI GPUs that can compete with NVIDIA.

According to market estimates, AMD's AI GPU share is still at around 5-8%, but the Instinct series such as MI350X and MI355X are gaining importance in terms of memory capacity, price, and procurement distribution.

According to AMD official information, the MI350X has 288GB of HBM3E memory and 8TB/s bandwidth, and supports AI and HPC applications with the ROCm software stack.

Memory capacity and cost efficiency are important, especially in large-scale models, making AMD a "realistic option for cloud companies to avoid relying solely on NVIDIA."

However, there are software barriers to completely replacing NVIDIA.

AMD is competing against CUDA with ROCm, but many AI developers have code, libraries, and operational know-how that are based on CUDA.

While that makes AMD a strong contender for NVIDIA's price negotiation cards, back-up in times of supply shortages, and cost optimization for specific applications, it still has a long way to go before it can usurp the standard across AI infrastructure.

2. Big Tech ASICs: The Biggest Structural Risk of the Inference Market

For NVIDIA, what could pose a greater structural threat than AMD is the ASICs developed by Big Tech customers themselves.

Representative examples are as follows.

The strength of ASICs is that they can improve cost and power efficiency by focusing on specific applications.

Google Cloud positions TPU v6e for Transformer, image generation, and CNN learning, fine-tuning, and serving.

AWS is rolling out Trainium as its own chip for AI learning, and is launching EC2 Trn2, which is based on Trainium2, for large-scale generative AI models.

Microsoft has also announced Maia 200 as an accelerator for AI inference, with the aim of improving cost per performance.

These aren't meant to completely replace NVIDIA's GPUs.

However, in inference applications where a large number of fixed models are run on a company's services, ASICs may be more rational than expensive general-purpose GPUs.

Therefore, NVIDIA's long-term risk is not that it will lose on learning GPUs, but rather that some of its inference will be absorbed into customers' own chips.

Why NVIDIA's sales are still growing

There are two reasons why NVIDIA's sales continue to grow despite increasing competition.

1. The market pie is bigger than that.

AI infrastructure investment is still in an expansion phase.

According to reports, the 2026 CapEx plans of major hyperscalers such as Microsoft, Alphabet, Amazon, and Meta are expected to grow to a total of around $700 billion.

In this market, even if NVIDIA's share declines a little, the absolute amount of sales can increase if the overall market expands.

In other words, investors should look at more than just market share.

NVIDIAKey point
AIinfrastructuremarket：Key point
Key point

As long as this pattern continues, the rise of AMD and ASIC will not be the end of NVIDIA, but will also be an indicator of the huge growth in demand for AI computing.

2. Supply constraints for advanced manufacturing and packaging

Cutting-edge AI chips require not only design, but also TSMC's advanced processes, advanced packaging such as CoWoS, HBM supply, and server assembly capabilities.

There is a strong perception in the market that NVIDIA is preferentially securing TSMC's advanced packaging capabilities.

Even if AMD, Broadcom, Marvell, Google, and Amazon design great chips, there are supply constraints on how many they can actually mass produce.

This point is also the reason why NVIDIA's market share is unlikely to collapse suddenly.

Perspective on Japanese stocks

What is important for investors is that the AI semiconductor market is moving from being dominated by NVIDIA to a hybrid era of GPUs and in-house ASICs.

While this means there is a risk that NVIDIA's growth rate will eventually slow down, it is not necessarily a bad news for Japanese semiconductor manufacturing equipment and parts companies.

This is because, whether the chips being manufactured are NVIDIA GPUs, AMD GPUs, Google TPUs, Amazon Trainium, or Microsoft Maia, if the total amount of cutting-edge chips increases, demand for manufacturing equipment, inspection equipment, materials, cutting/polishing, and packaging will likely increase.

Among Japanese stocks, Lasertec, Tokyo Electron, DISCO, Advantest, SCREEN, IBIDEN, AI server components, etc. are likely to be seen in the context of the ``total increase in AI chip volume'' rather than NVIDIA alone.

In other words, the rise of competing chips is a medium- to long-term risk for NVIDIA, but for Japan's semiconductor supply chain, it may actually mean an expansion of investment themes.

NVIDIA's real selling point towards 2027

The most important thing about these financial results is whether growth will continue until 2027.

NVIDIA's strength lies not only in its immediate Q1 numbers. It is on the roadmap connecting Blackwell, Rubin, agent AI, inference demand, and AI Factory.

1. $1 trillion demand for AI

At GTC in March 2026, multiple media outlets reported that CEO Jensen Huang expressed the view that Blackwell and Rubin's AI chip sales opportunity could reach $1 trillion by 2027.

What I would like to note here is that this should not be read as simply confirmed sales or an accounting backlog.

From an investor's perspective, it is appropriate to view this in the sense that NVIDIA itself has an extremely strong vision of AI infrastructure demand until 2027.

2. Roadmap from Blackwell to Rubin

NVIDIA has already begun mass production and shipment expansion of the Blackwell generation.

The next focus is on Rubin.

Officially, the Rubin platform is said to reduce inference token costs by up to 10x compared to Blackwell.

This is very important.

This is because, as the main field of generative AI moves from model learning to inference, agent AI, and robotics, cost efficiency and power efficiency will become more important for corporate adoption.

3. Agent AI will boost demand for inference

From the second half of 2026 to 2027, market interest will shift to agent AI.

Agent AI is not just an AI that answers questions, but an AI that autonomously executes tasks.

In this case, the demand does not end with one-time learning.

As AI continues to operate 24 hours a day, searching, determining, executing, recording, and relearning, the demand for inference will continue to grow.

If this structure continues, NVIDIA's growth drivers will expand from "learning GPUs" to "entire inference infrastructure."

3 stock price scenarios after settlement of accounts

As of May 19, 2026, NVDA stock is valued at around $220.10, with a market capitalization of approximately $5.39 trillion.

In the options market, price movements are priced in around 6-8% during the week of earnings announcements. OptionSlam puts the weekly implied move at around 7.5%, while Options Trading Report puts it at around 7.15%.

In other words, the market capitalization can change by tens of trillions of yen in a single day, whether the results are positive or negative.

Scenario A: Bullish break

In this case, the stock price could attempt an 8-12% rise after the earnings call.

The price is expected to rise above the recent high of $236.54 and test the low $240s.

Associative buying is likely to spread not only to AI semiconductors, but also to power, cooling, servers, semiconductor manufacturing equipment, and Japanese AI-related stocks.

Scenario B: Flat even with good financial results

In this case, the stock price may fluctuate wildly after hours, but eventually settle down to around ±2%.

NVIDIA has already factored in high expectations, so it is easy to judge that it has been ``priced in'' based on normal strong financial results.

In this scenario, we will be waiting for the progress of Rubin, Sovereign AI, and Agent AI as the next step.

Scenario C: Drops sharply due to unmet expectations

In this case, the stock price could test a decline of around 5-8%.

Technically, the price will first be confirmed at around $210, and below that, the price will see a downside confirmation zone around $193 to $195.

However, even in this scenario, unless demand for Blackwell and Rubin collapses toward 2027, the medium- to long-term growth story will not be completely destroyed.

Points that investors should really look at

The following eight points should be noted in this NVIDIA financial results.

The strong Q1 results have been factored into the market to some extent.

The real focus is whether this growth rate can be sustained until 2027.

Impact on Japanese stocks

NVIDIA's financial results have a direct impact on the Japanese market.

The reason is simple.

This is because NVIDIA is a huge ordering source at the top of the semiconductor supply chain.

If NVIDIA says it will ``increase production of Blackwell,'' ``proceed with the launch of Rubin,'' and ``demand for AI Factory is strong,'' funds will easily flow to TSMC, which is responsible for manufacturing it, and to the Japanese companies that supply TSMC with equipment, materials, and testing technology.

Supply chain structure

NVIDIA is a fabless company and does not have its own semiconductor factory.

As a result, the manufacturing, advanced packaging, testing, and assembly of cutting-edge GPUs relies on external supply chains.

The structure can be simplified as follows.

Microsoft、Google、Amazon、MetaKey point
↓
AIinfrastructureinvestment、GPUKey point
↓
NVIDIA
↓
GPUKey point
↓
TSMC
↓
3nm / 2nm、CoWoS、Key point
↓
JapanKey point

This is where Japanese stocks become important.

NVIDIA's AI chips are manufactured through multiple processes such as EUV exposure, advanced processing, CoWoS, HBM, inspection, cutting, polishing, and mounting.

Japanese companies' equipment and materials are used in each process.

Main related Japanese stocks

Lasertec has established itself as the de facto standard for EUV mask blank inspection equipment.

Tokyo Electron is strong in coater/developers used before and after EUV exposure, and the company's materials indicate a high market share.

Advantest is easily linked to demand for pre-shipment testing of AI semiconductors.

DISCO is notable for its thinning, cutting, and grinding processes that are important in advanced packaging and high-performance chips.

In other words, the stronger NVIDIA issues guidance, the more likely it is that Japanese equipment, inspection, materials, and parts companies will expect to receive more orders from next fiscal year onwards.

Why do stock prices move in real time?

There is a time lag in the reflection in business results.

Even if NVIDIA raises its demand forecast, there will be a lag of several months to more than a year before it is reflected in the sales of Japanese equipment manufacturers due to the process of ordering, manufacturing, delivery, and acceptance inspection.

Still, stock prices move quickly because the stock market anticipates future orders.

NVIDIAQ2
↓
TSMCKey point
↓
JapanKey point
↓
Key point

On the other hand, if NVIDIA does not meet expectations, there will likely be profit-taking selling in semiconductor-related stocks in the Japanese market the next business day.

This is not due to a sudden deterioration in the performance of individual companies, but rather due to a temporary decline in expectations for the AI ​​semiconductor cycle as a whole.

Impact of ETFs and theme funds

Another important thing is the existence of ETFs and investment trusts.

Institutional investors around the world often buy and sell the AI ​​and semiconductor sectors as a single theme.

Therefore, when NVIDIA stock moves significantly, there will likely be inflows and outflows of funds to the SOX index, semiconductor ETFs, and Japanese semiconductor-related stocks.

Companies such as Tokyo Electron, Advantest, and Lasertec are easily viewed by foreign investors as "Japan's AI semiconductor supply chain."

This is why stocks are mechanically bought and sold in the wake of NVIDIA's financial results, even without material from individual companies.

Points that investors should look at

EPS is not the only thing Japanese stock investors should look at in NVIDIA's financial results.

Rather, the following comment is important.

NVIDIA is one of the biggest engines for the Japanese stock semiconductor sector.

If the engine accelerates, TSMC, Japanese equipment stocks, inspection stocks, and parts stocks are likely to be pulled in as well.

Conversely, if the engine slams on the brakes, profit-taking will spread throughout the supply chain.

The reason why this financial statement is important for Japanese stocks is precisely because of this consolidated structure.

Conclusion

NVIDIA is a company at the center of the AI market.

However, stock prices are already factoring in one of the world's strongest growth rates.

What is important in this financial result is not whether Q1 sales are good or not.

What we really need to see is Q2 guidance, Blackwell's supply, Rubin's progress, and how realistic the $1 trillion AI demand will be until 2027.

If strong numbers and strong guidance come together, NVIDIA could once again push the overall AI market higher.

On the other hand, if there is even a slight slowdown in growth or supply delays, profit-taking will increase due to the huge market capitalization.

The most important thing for investors is to confirm whether NVIDIA will continue to be the center of AI infrastructure toward 2027, without being swayed by short-term price movements after the financial results.

Reference information

• NVIDIA Sets Conference Call for First-Quarter Financial Results
• NVIDIA Announces Financial Results for Fourth Quarter and Fiscal 2026
• S&P Global Market Intelligence: Nvidia earnings preview Q1 2027
• TrendForce: HBM4 Mass Production Delayed to End of 1Q26
• TrendForce: Rubin Faces Delay Risks, Blackwell to Account for Over 70% of High-End GPU Shipments
• The Register: Supply chain challenges risk delaying Nvidia's Rubin GPUs
• Motley Fool: Nvidia Earnings on May 20
• Reuters via Investing.com: Nvidia sales opportunity for Blackwell, Rubin chips more than $1 trillion by 2027
• Tom's Hardware: China is blocking Nvidia H200 purchases despite US approval
• TechInsights: The Inference Bottleneck
• NVIDIA Developer: About CUDA
• NVIDIA AI Enterprise
• NVIDIA Omniverse and Cosmos for physical AI
• AMD Instinct MI350X GPUs
• Google Cloud TPU v6e / Trillium
• AWS Trainium
• Microsoft Maia 200
• Tom's Hardware: Big Tech AI capex spending plans
• Lasertec: Business and Core Technology
• Tokyo Electron: Corporate Update
• TSMC 4Q25 Management Report / 2026 CapEx outlook
• OptionSlam: NVDA earnings implied move
• NVDA market data