Are high-tech giants (including Deepseek) calling nvidia “checkmates”?

nvidia (NVDA) 2.63%)) So far, he has been a clear winner in the AI ​​revolution. The company has become a major AI chip designer and owns approximately 80% of the market. Nvidia creates the world’s most powerful graphics processing unit (GPU). These are important chips for key AI tasks such as model training and guessing.

The biggest technology companies have gathered at Nvidia to take the power of AI projects for these sought-after GPUs. This has resulted in double- and triple-digit revenue growth for NVIDIA after the quarter. And this revenue has reached record levels, with NVIDIA reporting more than $35 billion in the most recent period. On top of this, Nvidia is extremely profitable in sales, with a continuous total margin of over 70%.

However, one particular risk has recently attracted investors’ attention. This is a risk that can intensify and undermine revenue growth. It links to actions taken by some of our Nvidia customers. Do these tech giants say “checkmate” to nvidia? Let’s look into it.

Image source: Getty Images.

Nvidia’s most powerful chip

Let’s start by talking about Nvidia and its customers. Nvidia sells a variety of products and services to AI customers, but the focus is on the company’s GPU. As mentioned before, they are the most powerful, with $30,000 pop and the most expensive – and Nvidia has promised to renew them every year. This is a move that allows chip designers to move ahead in terms of product technology and quality.

Nvidia has not revealed the names of its biggest customers, but we can identify them by considering comments from some of the world’s leading tech companies. Companies from Amazon to Meta Platforms and Openai have spoken about using Nvidia chips. For example, a year ago, Meta predicted that by the end of 2024, it would have 350,000 Nvidia H100 GPUs.

Another example: Oracle co-founder Larry Ellison said a few months ago that he and Tesla chief Elon Musk “begged” Nvidia for more tips.

So it is clear that since the start of the AI ​​boom, the tech giant has been turning to Nvidia to power the project. Now let’s consider some of the moves that could represent the risk of chip leaders. It has to do with competition, but not from rivals we expect, such as fellow chip designers Advanced Micro Devices or Intel. Instead, the competition comes from Nvidia’s clients’ research and development department.

Meta’s “pursuit of cost efficiency”

Many people produce their own chips as a way to reduce their reliance on NVIDIA or as a way to provide customers with access to low-cost chips. Amazon develops a chip training line and sells it to cost-conscious customers through Amazon Web Services Business. In its recent revenue report, Meta said it would “pursuit of cost efficiency” by deploying MTIA chips in rankings and recommended inference workloads for ADS and organic content.

And recently, Reuters reported that Openai has finished its own AI chip design and is hoping to begin manufacturing phases in the coming months.

Finally, last month, a report from Chinese startup Deepseek reported that it trained models with Nvidia’s low-performance chips suggested that companies may not always need to invest a large amount of money to get the job done. It is reported that there is a possibility of strengthening.

Given these moves by the tech giants, it may seem like they are saying “checkmates” to nvidia – revenue growth for chip leaders as these players turn to low-cost chips I’ll delay it. They were developed at home.

Is the chess game over?

However, before flipping through this chess game, it is important to consider the big picture before declaring Nvidia as a loser. Yes, these NVIDIA customers are expanding their chip options. However, I don’t necessarily think this will hurt Nvidia’s growth. This is why. Firstly, the tech giants are looking for excellence, so far nvidia is looking for the best to drive their AI program. Therefore, they can play a role in deploying Nvidia’s latest GPU for the most important tasks and supporting some of their own chips. And building on current NVIDIA systems is the number one interest of our customers to reduce the total cost of ownership over time.

It is also important to remember that Nvidia does not sell not only GPUs, but the entire ecosystem of products and services to advance AI projects. Therefore, the company’s revenue is not solely dependent on GPU. And as the AI ​​revolution moves to the next stage, Nvidia’s revenue opportunities could actually expand. For example, more companies may turn their attention to Nvidia’s enterprise software products.

All this means that Nvidia has a lot of potential moves for the Chess Committee, allowing us to ensure revenue growth and share performance as this AI boom continues.