Nvidia (NVDA 0.38%) released its fiscal 2027 first-quarter results (for the three months ended April 26) on May 20, and the company’s numbers make it clear that its grip on the artificial intelligence (AI) chip market remains as strong as ever.
Nvidia has dominated the AI chip market since the technology gained prominence in November 2022, following the launch of ChatGPT. Since then, companies such as Intel (INTC +3.02%), Advanced Micro Devices (AMD +7.91%), and others have tried to make a dent in the AI chip market. While there is no denying that Nvidia’s competitors have succeeded to some extent, the semiconductor specialist’s phenomenal growth and product strategy have helped it stay ahead of the curve.
The good news for Nvidia stock investors is that the company isn’t resting on its laurels. Comments by management on the company’s latest earnings call make it clear that it will try to dominate every nook and cranny of the AI chip market, which could be bad news for Intel and AMD. Let’s look at the reasons why.
Image source: The Motley Fool.
Nvidia’s server CPU business is going to take off this year
Nvidia’s graphics processing units (GPUs) have been the foundation of the company’s AI chip empire. These chips are capable of performing massive parallel calculations, which is why they have been preferred over the years for training large language models (LLMs). As AI compute is now moving toward inference, Nvidia is tuning its GPUs to perform inference applications as well.
Today’s Change
Current Price
However, it’s worth noting that the shift toward inference has also created a need for energy-efficient processors. That’s because the inference phase is less compute-intensive, which is why it doesn’t need the massive computing power offered by GPUs. As a result, server central processing units (CPUs) have been in great demand of late, fueling healthy growth in AMD and Intel’s AI revenue.
AMD and Intel are the leading players in the server CPU market. They design and manufacture chips using the x86 architecture. Both companies have recently reported that hyperscalers have been deploying their server CPUs in AI data centers to run inference applications. AMD saw robust demand for its Epyc server processors in Q1, which led to a 57% year-over-year increase in its data center revenue to $5.8 billion.
Meanwhile, demand for Intel’s Xeon server processors is exceeding supply, which was a key reason its data center and AI (DCAI) revenue increased by 22% year over year in Q1 to $5.1 billion. Nvidia has been missing out on this market so far, as it doesn’t offer a stand-alone server CPU.
Though the company does offer a server CPU, known as Grace, it pairs it up with its GPUs and other chips as a part of a rack-scale system. However, Nvidia now wants a piece of the stand-alone server CPU market. That’s not surprising as AMD recently doubled its server CPU total addressable market (TAM) estimate to $120 billion by 2030.
Nvidia believes that this opportunity is much bigger. It sees the server CPU market unlocking a $200 billion revenue opportunity for it in the long run. Importantly, Nvidia has decided to offer its next-generation Vera server CPU as a stand-alone product to customers, and the company anticipates $20 billion in revenue from it in 2027. CFO Colette Kress remarked on the earnings call:
Vera CPU opens a brand-new $200 billion TAM for Nvidia, a market we have never addressed before. Every major hyperscale and system maker is partnering with us to get it deployed. We have visibility to nearly $20 billion in total CPU revenue this year, setting us up to become the world’s leading CPU supplier.
This statement clearly indicates that Nvidia is well on its way to dominating the server CPU space.
Why Nvidia’s foray into server CPUs is bad news for Intel and AMD
AMD noted in November 2025 that the server CPU market was estimated at $26 billion in 2025. Nvidia’s $20 billion server CPU revenue projection suggests that it is poised to make a huge dent in this market this year. Even if the server CPU market were to double in 2026 and Nvidia indeed achieved its $20 billion goal, it would still be able to corner almost 40% of this lucrative space.
An important point worth noting is that Nvidia utilizes Arm Holdings‘ architecture to design its server CPUs. Hyperscalers have been preferring Arm’s architecture over Intel and AMD’s x86 architecture to reduce costs and improve computing efficiency. This is why Counterpoint Research expects Arm-based server CPUs to account for a whopping 90% of the AI data center market by 2029. That would be a massive jump, considering that Arm reportedly controlled 15% of the data center CPU market a year ago.
With backing from a major chip designer such as Nvidia, it won’t be surprising to see the Arm architecture dominate server CPUs. So, Nvidia could eventually eat Intel and AMD’s lunch in the server CPU market. Throw in Nvidia’s dominant position in GPUs, and it is easy to see why it seems like the best semiconductor stock you can buy to capitalize on the AI chip market’s growth, especially considering that it trades at just 25 times forward earnings.
