After a 20% correction in Nvidia's stock price, is its moat being eroded from cloud services to open

Nvidia (NVDA) has become an investor favorite thanks to its dominant position in the field of artificial intelligence (AI). However, although its financial performance remains strong, recent market reactions have sparked widespread concern and worry. Especially after the release of Nvidia’s Q2 2024 earnings report, Nvidia’s stock price once fell by nearly 20%, even though the company exceeded its expected revenue and earnings per share (EPS) targets. Although the stock price has partially rebounded since then, the market seems to have greater doubts about future profitability and risks on the demand side.

Nvidia’s growth largely depends on its data center business, especially the investment in AI infrastructure by large cloud service providers such as Amazon AWS, Google Cloud, and Microsoft Azure.

Nvidia’s chief financial officer, Colette Kress, mentioned in a recent earnings conference that “about 40-50% of our data center revenue comes from these cloud service providers.” However, as the capital expenditures (Capex) of these tech giants peaked in recent quarters, the market has uncertainties about whether they can continue to invest significantly in the future. The possibility of such a slowdown in investment may pose significant pressure on Nvidia’s long-term revenue growth.

Nvidia’s risks seem to be growing day by day.

Nvidia (NVDA) has achieved significant growth in recent years thanks to its technological advantages and market dominance in the field of artificial intelligence (AI). However, recent market performance shows that investors are worried about its future profitability and the sustainability of demand.

In the latest earnings report, Nvidia reported strong revenue growth, especially the outstanding performance of the data center business. According to the disclosure of Nvidia’s chief financial officer Colette Kress, about 40-50% of the data center revenue comes from large cloud service providers such as Amazon (AMZN), Google (GOOGL), and Microsoft (MSFT). These tech giants have significantly increased capital expenditures in recent years, mainly to support the expansion of AI infrastructure. However, as the latest earnings reports of these companies show that profit margins and revenue growth are lower than expected, the market is worried that future capital expenditures may slow down, which will put pressure on Nvidia’s revenue growth.

Although Nvidia dominates the AI chip market, its competitor AMD (AMD) is rapidly expanding its AI business. The latest data shows that AMD expects its AI chip revenue to reach $4.5 billion in 2024, indicating that market competition is intensifying. This rapid growth may pose a greater challenge to Nvidia’s market share and profit margin.

In addition, Nvidia’s future earnings expectations are not completely optimistic. Analysts predict that by fiscal year 2027, Nvidia’s earnings per share will reach $4.71, corresponding to a forward price-earnings ratio of 26.2 times.

Such a high valuation means that any decline in profit margins or weak demand may lead to significant fluctuations in stock prices. Investors can use the multi-asset wallet BiyaPay to regularly check Nvidia’s market trends and choose trading opportunities.

If there are problems with deposit and withdrawal, you can also use it as a professional tool for US and Hong Kong stock deposit and withdrawal. You can recharge digital currency and exchange it for US dollars or Hong Kong dollars, withdraw it to a bank account, and then deposit it into other brokers to buy stocks. The arrival speed is fast and there is no limit, so you won’t miss investment opportunities.

Although Nvidia has strong short-term financial performance, investors had better closely monitor the capital expenditures of large cloud service providers, changes in the market competition pattern, and the sustainability of the company’s profit margins. These factors will directly affect Nvidia’s performance in the AI market in the future.

Main risks

Although Nvidia (NVDA) has achieved great success thanks to its leadership position in the field of artificial intelligence (AI), the current risks cannot be ignored. Especially under the threats of a slowdown in capital expenditures, the rise of competitors, and the potential threat brought by open-source technology, Nvidia’s growth prospects face many challenges.

Slowing investment and demand from cloud service providers

A large part of Nvidia’s revenue depends on its data center business, especially large investments from cloud service providers such as Amazon (AMZN), Google (GOOGL), and Microsoft (MSFT). These tech giants have significantly increased capital expenditures in recent years, driving Nvidia’s revenue growth. Nvidia’s chief financial officer Colette Kress revealed that about 40-50% of Nvidia’s data center revenue comes from these cloud service providers.

However, there is uncertainty in the market about whether these companies can continue to maintain such high-level expenditures in the future.

To measure Nvidia’s growth potential, we need to look at the recent performance of some of its largest customers. Most senior executives of large cloud providers are very optimistic about their AI services, but Wall Street has doubts about their future forecasts. We have seen this in the recent earnings season. The performance of large cloud players is not good.

In the first quarter’s earnings, Nvidia’s chief financial officer Colette Kress mentioned that large cloud providers contributed a percentage in the mid-40s to Nvidia’s data center revenue base.

If these large cloud providers are strongly opposed by Wall Street due to their AI expenditures, it will inevitably damage Nvidia.

The three major cloud providers performed poorly in this earnings season. All of them are clearly lagging behind the broader market. They have company-specific issues, but one of the common bearish arguments for them is the large amount of capital expenditures on AI, which has not brought improvement in profit margins and revenue trajectories.

We can look at the revenue trend of Amazon’s AWS in the past few quarters.

AWS has been investing heavily in AI, hoping to launch new AI services and attract more customers. However, this has not significantly increased the company’s revenue trajectory. Similar trends can be seen in Google Cloud and Microsoft’s Cloud.

In the past few quarters, the capital expenditures of the three major cloud providers have shown a clear inflection point. Now each of them estimates that the capital expenditure for this fiscal year is close to $50 billion. Most of the incremental expenditures are used to build AI services and purchase expensive chips from Nvidia. However, if we do not see substantial improvement in the profit margins and revenues of these cloud players, Wall Street may become very pessimistic about these investments.

Nvidia’s management is betting that AI demand will continue to increase in the next few quarters. However, when we look at macro charts and the performance of some major customers, this is not guaranteed.

Increasing competition from AMD

Although Nvidia has been dominant in the AI chip market, its competitor AMD has continuously increased its investment in the field of AI chips in recent years, gradually forming a situation to compete with Nvidia. AMD’s AI chip revenue is expected to reach $4.5 billion in 2024, much higher than expected in previous years. This rapid growth will pose a threat to Nvidia’s market share and profit margin.

AMD expects its AI chip revenue to reach $4.5 billion in 2024. This estimate has been continuously rising in the past few quarters.

At the end of 2023, AMD estimated that AI chip revenue was only $2 billion. The final result in 2024 is likely to easily exceed the $5 billion mark. Although this is only a small part of Nvidia’s revenue base, we can see an obvious trend. If AMD’s current growth trajectory continues in 2025, we may see significant resistance to Nvidia’s profit margin. Although Nvidia has a higher market share, Wall Street may become more and more cautious about the company’s ability to maintain particularly high profit margins.

In the most recent quarter, revenue has declined consecutively. Nvidia’s revenue in the United States was $13.5 billion in the first quarter of 2025 and $13 billion in the second quarter of 2025.

A possible explanation for this sequential decline is that large technology customers have begun to reevaluate their AI expenditures. The possibility that this sequential decline will continue in the next few quarters is very high, which will have an adverse impact on the company’s revenue.

Increasing risk of open-source challenges to Nvidia CUDA

The rise of open-source technology is another important factor that poses a threat to Nvidia’s future growth. In recent years, technology companies have increasingly adopted open-source AI solutions to reduce their dependence on proprietary technologies.

In September 2023, a group of tech giants gathered to establish the Unified Accelerator Foundation (or “UXL Foundation”) based on Linux infrastructure.

The main goal of this foundation is essentially to create an open-source alternative to Nvidia’s CUDA software platform. Some of the main founding members of this plan include Alphabet Inc. (GOOG), Google Cloud (GOOGL), Arm Holdings plc (ARM), Intel Corporation (INTC), Qualcomm Incorporated (QCOM), Samsung, and VMWare, which is now owned by Broadcom Inc. (AVGO).

Our members are united in support of open standards and open source to build a multi-architecture and multi-vendor software ecosystem for all processors, including CPUs, GPUs, and AI processors. ——UXL Foundation

Key customers like Google leading the effort to create an alternative to CUDA is a force to be reckoned with. This is especially true given that among all three major cloud providers, Google has the strongest custom chips to handle generative AI workloads.

Thanks to this leading position, it is most capable of building its own software ecosystem around internal AI processors.

Regardless of the leading position among cloud providers, the fact that Google contributes its expertise to the development of this open-source software shows that the company is unwilling to continue to pay high prices for Nvidia’s GPUs.

And the UXL Foundation is not the only effort to address Nvidia’s CUDA moat.

The UXL Foundation’s plan is one of many efforts to weaken Nvidia’s control over AI software.

According to custom data compiled by PitchBook at the request of Reuters, investors and corporate funds have invested more than $4 billion in 93 independent projects.

Open source essentially means that the code is completely transparent, which basically allows any third-party developer to customize and build software, enabling faster progress and innovation.

At the same time, Nvidia is still investing heavily in internal developers to innovate new products and services and expand the functionality of its platform.

On a sequential basis, GAAP and non-GAAP operating expenses increased by 12%, mainly reflecting an increase in compensation-related costs.

As we strive to develop next-generation products, operating expenses for the full year are expected to increase in the upper middle range of 40%.

——Chief Financial Officer Colette Kress, Nvidia’s Q2 2025 earnings conference call

Innovating increasingly powerful AI solutions does incur higher operating expenses. But Nvidia cannot charge a corresponding premium for its Blackwell product to cover the increased costs of developing such powerful AI technology. As competitors’ chips catch up and become compelling alternatives, and the rise of open-source software erodes Nvidia’s biggest moat, the company’s pricing power is likely to further deteriorate.

Therefore, the dual effects of slowing revenue growth and shrinking profit margins may dampen the current excitement around Nvidia and ultimately likely lead to a significant correction in stock prices.

Nvidia’s moat is not limited to CUDA

Now, although the threat of open-source software is growing, note that CUDA is not the only software service provided by Nvidia.

Among the many different software services owned by Nvidia, perhaps the most promising is the “NVIDIA AI Enterprise” operating system. In fact, Nvidia is well-prepared for this AI revolution and has even developed its own operating system to run complex AI workloads.

Just as the rise of Amazon’s Linux-based AWS subsequently gave rise to new software-as-a-service (SaaS) companies, it is worth noting that several of these SaaS companies have become important customers of Nvidia’s AI operating system.

In fact, software giants such as ServiceNow, Inc. (NOW) and Adobe Inc. (ADBE) have been among the earliest adopters of “NVIDIA AI Enterprise”. Although Salesforce, Inc. (CRM) has been mainly relying on Google Cloud TPU to train its own AI models, this enterprise software giant finally revealed a partnership with Nvidia last week to build an AI agent.

Now there is no doubt that just as the UXL Foundation is working hard to create an open-source alternative to CUDA, efforts will also be made to develop an open-source competitor to “NVIDIA AI Enterprise”.

What we will witness in the next ten years is a tug-of-war: Nvidia will strive to deeply integrate customers into its ecosystem through enhanced value propositions, while open-source providers will seek to attract customers by simplifying the conversion process and announcing advantages such as cost-effectiveness and code transparency.

Through software services, Nvidia is gradually transitioning from traditional hardware sales to the SaaS model.

Unlike one-time hardware sales, software services can bring a continuous revenue stream for Nvidia. Customers can pay regularly by subscribing to platforms such as Nvidia AI Enterprise to obtain continuous service support and technical updates. This SaaS model brings revenue diversification for Nvidia and reduces its dependence on hardware sales.

According to Nvidia’s strategic plan, the company plans to continue to expand its software and service business in the next few years. This expansion can not only help the company find new revenue sources outside the hardware market but also improve its overall gross margin and profitability. By providing an integrated AI hardware and software solution, Nvidia is gradually becoming a comprehensive service provider in the global AI market.

Not only that, once an enterprise’s AI/LLM (large language model) is trained on Nvidia’s proprietary AI software platform CUDA, considering the cost, it is almost impossible to transfer to other platforms. As the LLM further develops on this platform in the next few years, this moat is expected to become even stronger.

Based on such a trend, it is predicted that Nvidia’s market share may slightly increase to 90% within ten years. This prediction is positively affected by the upcoming Rubin chip series, which provides a solid foundation for Nvidia to further launch new chip series. In addition, Nvidia is using its own AI technology to design chips, which other companies cannot achieve at present. Nvidia can also use its Blackwell chip to perform the same task and even deploy it before other companies obtain Blackwell.

In short, Nvidia can use the latest AI chips and software to design better next-generation chips and do it earlier than other competitors. This is one of the reasons why many people think Nvidia has a deep moat and part of the reason why it has an AA- credit rating.

Is it still worth buying now?

From an optimistic perspective, Nvidia’s long-term market opportunity is still huge.

Especially the market demand brought by data center upgrades may reach a scale of $2 trillion, further supporting Nvidia’s leadership position in the field of accelerated computing platforms. Nvidia’s strong technological advantages in AI infrastructure and ecosystem integration beyond the CUDA platform mean that it will not be easily surpassed by competitors in the short term.

Even under bear market expectations, its downside is limited, and the upside potential is still considerable.

For investors who already hold Nvidia stocks, considering Nvidia’s technological advantages and future market potential, it is a reasonable strategy to continue holding the stock despite facing market competition and the threat of open-source technology.

However, for investors who do not yet hold Nvidia positions, the current high valuation may not be suitable for immediate entry. With the risks of slowing revenue growth and declining profit margins, the possibility of a stock price correction still exists in the short term.

Therefore, investors should carefully decide whether to increase holdings or buy Nvidia stocks for the first time according to their own risk tolerance.