The AI King Has Competition

On June 13, Nvidia became the first chip maker to join the “$1 Trillion Club.” That was the day that the stock’s market capitalization exceeded $1 trillion. This put Nvidia in elite company alongside Amazon, Microsoft, and Apple.

And it was all thanks to its H100 AI chips.

Nvidia holds roughly 90% of the AI chip market today. It’s forecasting 50% growth in sales growth over the next quarter.

The AI chip market is expected to grow from roughly $30 billion this year to $150 billion by 2027.

Nvidia is the obvious AI play to capture that growth. And that’s why investors rewarded the company with a $1 trillion market cap. There’s just one problem.

The company’s stranglehold on the market won’t last.

There are three threats to Nvidia’s dominance developing today. But these competitors haven’t made their big move yet.

That’s an opportunity for us as investors to get in early before the word is out.

Let me explain.

Coming for The King

AMD is set to start shipping the MI300 chips this fall.

It’s AMD’s answer to Nvidia’s H100 chip which powers most of the AI data center industry today.

Once the MI300 hits the market, it’ll be the first real competitor to Nvidia.

AMD is showing off the chip’s strength by using it to power “El Capitan.” That’s the newest supercomputer from Lawrence Livermore National Lab, which is set to be unveiled this year.

It’s expected to take the record for the world’s fastest and most powerful supercomputer.

AMD’s MI300 is about a year late to the AI race. But its improved performance will allow it to fight Nvidia for market share.

But AMD isn’t the only one Nvidia needs to worry about.

Open-Source Tech is Gaining Traction

On Tuesday, I attended the Andes RISC-V Con 2023. It’s a free one-day conference on AI, data center, and open-source software.

The biggest complaint I heard from developers was Nvidia’s lack of customization. Nvidia’s GPUs are “closed-source” software. That means Nvidia designs it inhouse and won’t let users alter the code.

In some ways, this makes sense on Nvidia’s part. They basically own this market. There is no incentive for them to share their technology or address these concerns. After all, Nvidia is basically the only option.

But there’s a customizable alternative emerging.

It’s known as RISC-V (pronounced, risk-five). It’s an open-source instruction set for computer chips developed at University of California Berkeley in 2010. It’s free to use and expand upon. 

Until recently, RISC-V was mainly used by tech “nerds.” But some of the biggest tech companies are now moving to adopt it. 

In January, Google announced that RISC-V will be supported by Android. Google will be building specific software code for Android developers. Once completed, it will reduce the time it takes to develop applications that run on RISC-V.

Meta is making similar moves. In May, it announced a custom accelerator chip, the MTIA v1, based on RISC-V. The MTIA v1 handles AI tasks such as content understanding and ad ranking.

Development of AI will not happen exclusively behind the “walled garden” of the tech companies like Nvidia. RISC-V’s will give smaller developers and device makers the chance to build on innovative ideas.

On Shaky Ground

AMD is Nvidia’s biggest competitor… and RISC-V is the wild-card to watch.

But that doesn’t mean other chipmakers aren’t racing to cash in on the AI trend.

Micron builds memory and storage systems that support AI servers.

Broadcom makes chips and network technology used in AI data centers. It estimates that AI-related tech will grow from 10% of its business last year to 25% next year.

TSMC builds chips for other companies. It has seen an increase in demand due to AI and called it a “megatrend for business growth.”

The chip market was valued at $574 billion in 2022. It’s expected to more than double to $1.3 trillion in 2029. 

That means it’s a great time to be invested in chipmakers.

On Tuesday, I’ll share my three favorite companies to own right now to gain exposure to the AI trend. And yes, one of them is AMD. I’ll share my full research next week.

But the important thing I’d ask you to understand today is that Nvidia’s dominance will not last. And that’s a problem for anybody buying the stock today.

NVDA trades at around 40 times sales as I write this. The stock has never traded this expensively on a valuation basis. Maybe Nvidia can “grow into” that valuation. But for all the reasons I shared above, I’m not willing to take that bet.

The smart move is to step aside from NVDA for the time being. Let’s explore the chipmakers and AI adopters that investors aren’t looking at. As I said, I’ll share my top three AI stocks to own next week.

Regards,

Colin Tedards
Editor, The Bleeding Edge