Generative artificial intelligence (AI) uses large language models and other machine learning models to create text, images, video, audio, and computer code. Use cases range from digital assistants that improve worker productivity to intelligent avatars that make video games more lifelike.
Bloomberg Intelligence expects generative AI revenues to top $1.3 trillion in 2032, up from $63.5 billion in 2023. In other words, Bloomberg expects spending on generative AI hardware, software and services to increase by 2,040% over the next nine years.
Some analysts see an even bigger market. Research from McKinsey & Company, for example, suggests that generative AI will eventually contribute $7.9 trillion annually to the global economy. Such opportunities come along infrequently, so investors should position their portfolios accordingly.
Nvidia (NASDAQ: NVDA) is a logical stock to own. In fact, I think most investors should have some exposure to the chipmaker. But Nvidia is far from the only company that stands to benefit, and owning a single AI stock is a bad strategy. Here’s why I think Supermicrocomputer (NASDAQ: SMCI) is the best AI stock to buy now.
Nvidia dominates the artificial intelligence processor market
Nvidia graphics processing units (GPUs) have long been the gold standard for accelerating data center workloads like artificial intelligence. Attention has certainly increased over the past year, as Nvidia’s revenue and net income have grown at triple-digit rates for four consecutive quarters. But investors should understand that Nvidia hasn’t stumbled into its recent success without a fight.
Nvidia has dominated the data center GPU market for nearly two decades, and the company has been setting performance records on the MLPerfs, objective benchmarks that measure how quickly AI systems can complete AI training and inference tasks, ever since it introduced these tests in 2018. One reason for that success is the CUDA programming language, which allows Nvidia GPUs (originally built for computer graphics) to accelerate other data center workloads.
Nvidia introduced CUDA in 2006, and the platform has since expanded to include hundreds of frameworks and software libraries that streamline data preparation, model training, and AI application development. No other chipmaker has a supporting software ecosystem that comes close to matching it, so naturally Nvidia has become the leader in AI processors. To quote The Wall Street Journal“Nvidia chips underpin all of the most advanced AI systems, giving the company an estimated market share of over 80%.”
The problem (if there is one) is valuation. Wall Street expects Nvidia to grow earnings per share by 33% per year over the next three to five years. Divide that number by the current price-to-earnings ratio of 74 and the result is a price-to-earnings-growth ratio (PEG) of 2.2. That’s not excessive. That multiple is actually a discount from the three-year average of 3.1, so risk-tolerant investors should consider buying a small position today.
However, I think Super Micro (also called Supermicro) is a better buy right now because the stock is much cheaper. At the same time, the company can benefit greatly from companies investing in generative AI hardware.
Supermicro Gains Market Share in Artificial Intelligence Server Market
Supermicro builds accelerated computing platforms. Its portfolio ranges from individual servers and storage systems to full-rack solutions purpose-built for enterprise and cloud data centers. The company works closely with partners such as Nvidia, IntelAnd Advanced micro devices to equip its hardware with the latest chips.
Samik Chatterjee at JPMorgan Chase sees Supermicro as the “leading company in the AI computing market.” More importantly, Supermicro is quickly gaining market share. The company accounted for 10% of AI server sales in the quarter ended December, but KeyBanc’s Tom Blakely says that figure could reach 23% this year. He also believes Supermicro “has competitive moats that should maintain or even expand that share in the coming years.”
Blakely points to internal manufacturing capabilities and a unique building-block approach to product development. Engineers make up about half of Supermicro's workforce, and the company does most of its research and development in-house (in Silicon Valley). In turn, Supermicro can bring new products to market quickly and efficiently, and the modular approach only amplifies that ability.
Supermicro is often first to market with new technologies in particular because it can “quickly assemble a broad portfolio of solutions by leveraging common building blocks across all product lines.” In other words, the company can quickly equip partially pre-assembled servers with the latest central processing units (CPUs), GPUs, memory, and interconnects, often beating its competitors to market.
CEO Charles Liang recently emphasized that advantage: “We deliver optimized AI solutions at scale, giving us a time-to-market advantage and shorter lead times than our competitors.”
Supermicro trades at a more reasonable valuation than Nvidia
According to 650 Group, annual AI server shipments are expected to increase sixfold between 2023 and 2028. Supermicro is well-positioned to capitalize on increasing demand given its market leadership, which is itself supported by in-house engineering capabilities and a unique approach to product development.
Wall Street expects Supermicro to grow earnings per share by 48% per year over the next three to five years. When that number is divided by its current valuation of 47 times earnings, the result is a PEG ratio of less than 1. To be clear, Supermicro is no hidden gem. In fact, it was the best-performing stock in the S&P 500 in the first half of 2024. But its valuation is very reasonable, especially compared to Nvidia's PEG ratio of 2.2.
Should You Invest $1,000 in Nvidia Now?
Before you buy Nvidia stock, here are some things to consider:
The Motley Fool Stock Advisor team of analysts has just identified what they think is the 10 best stocks for investors to buy now… and Nvidia wasn’t one of them. The 10 stocks that made the cut could deliver monster returns in the years to come.
Think about when Nvidia made this list on April 15, 2005… if you had $1,000 invested at the time of our recommendation, you would have $771,034!*
Stock Advisor offers investors an easy-to-follow blueprint for success, including portfolio building guidance, regular analyst updates, and two new stock picks each month. The Stock Advisor has service more than quadrupled the return of the S&P 500 since 2002*.
View the 10 stocks »
*Stock Advisor returns as of July 2, 2024
JPMorgan Chase is an advertising partner of The Ascent, a Motley Fool company. Trevor Jennewine has positions in Nvidia. The Motley Fool has positions in and recommends JPMorgan Chase and Nvidia. The Motley Fool has a disclosure policy.
Generative AI Sales Could Soar 2,040%: My Picks for the Best AI Stocks to Buy Right Now (Hint: Not Nvidia) was originally published by The Motley Fool
