Nvidia Dominates AI Silicon, But That Doesn’t Justify A $1 Trillion Valuation
Nvidia (NASDAQ: NVDA) stock surged by 24% in Thursday’s trading after the company posted a better-than-expected set of Q1 FY’24 results and provided Q2 revenue guidance that was well ahead of Street estimates. With the surge, Nvidia’s market cap is approaching the elite $1 trillion mark. Generative artificial intelligence (AI) has captured the imagination of the world following the launch of the viral ChatGPT chatbot late last year. Technology companies and developers have been scrambling to deploy generative AI into their applications and investors are counting on this to drive a windfall of sorts for Nvidia, whose server chips remain the go-to products for AI workloads. So is Nvidia’s stock surge warranted and is the stock a buy at current levels of around $380 per share? We don’t think so. Below, we take a look at Nvidia’s role in the AI space and why we believe the stock is overvalued.
Artificial intelligence workloads require a considerable amount of computing capacity, shifting the power balance away from central processing units made by the likes of Intel to Graphics processors, a market which Nvidia dominates. For perspective, advanced AI computers today have as many as eight GPUs but just one or two CPUs and this trend is only likely to scale through the computing industry as more applications use generative AI capabilities. Nvidia has a suite of chips tailor made for AI. For example, the company’s H100 GPU offers high-end computing capabilities focused on deep learning, machine learning, and other AI tasks and is optimized for data center environments. The Grace Hopper processors are custom-designed CPU chips developed by Nvidia to work in conjunction with GPUs for AI workloads. The company has been looking to develop an ecosystem of sorts around its AI tools, with its own programming languages, and software, which are helping the company to better lock in customers. Nvidia appears to be preparing well on the production front to handle the surging demand, indicating that it had procured substantially larger commitments from its suppliers. The strong demand and supply planning are being reflected in the company’s Q2 guidance, with revenue projected at $11 billion at the mid-point (translating into 60% year-over-year growth), compared to the Wall Street consensus of just $7.2 billion in sales.
While we have raised our price estimate for Nvidia from around $178 per share to about $256 per share, to account for the stronger outlook for the data center business, we think the stock looks overvalued. At the current market price of about $380 per share, Nvidia stock trades at about 30x forward revenue and about 80x forward earnings. This compares to the broader semiconductor industry average price-to-sales multiple of about 4.5x. Even Tesla stock didn’t trade at P/S multiples of these levels at its peak in 2020-2021. Nvidia’s high multiples are not supported by the company’s potential growth rates and the possible risks that Nvidia faces in the space. Consensus estimates for 2025 point to just about 25% revenue growth. While these numbers might be conservative, we don’t believe even slightly better growth rates will support a multiple this high. Although Nvidia remains in pole position on the market for AI chips, competition in the AI market could mount, with traditional chipmakers such as Intel and AMD potentially inventing to catch up in this space given the high stakes. Moreover, big tech players such as Google are also doubling down on AI and machine learning-related silicon. For example, Google’s Tensor Processing Units are specialized integrated circuits focused on machine learning. See our analysis on Nvidia Valuation: Is NVDA Stock Expensive Or Cheap? for more details on what’s driving our price estimate for NVDA stock. Also, see our analysis of Nvidia Revenue.
Returns | May 2023 MTD [1] |
2023 YTD [1] |
2017-23 Total [2] |
NVDA Return | 37% | 160% | 1323% |
S&P 500 Return | 0% | 8% | 85% |
Trefis Multi-Strategy Portfolio | -1% | 8% | 237% |
[1] Month-to-date and year-to-date as of 5/26/2023
[2] Cumulative total returns since the end of 2016