For years, Advanced Micro Devices (AMD) has been considered a distant rival to Nvidia, despite both having a near-duopoly of the GPU market.

Even though the company has been building chips for AI and data center use since 2018, it has continued to lag behind Nvidia in both hardware design and software development needed for high performance computing. 

However, over the past year, the company has slowly built itself up as a credible alternative to Nvidia in the GPU space and Intel in the CPU market.

Investor Essentials Daily:
Tuesday News-based update
Powered by Valens Research

The AI boom has been a boon for chip manufacturers due to hyperscale demand for high performance chips.

One of the early winners from this surge in demand was high performance chip designer Nvidia Corporation (NVDA), which saw its market cap balloon from $400 billion at the end of 2022 to $5 trillion today.

Nvidia has largely maintained its lead in the past few years, as the company currently owns 86% of the AI data center chips segment. 

Despite Nvidia’s large lead, Advanced Micro Devices (AMD)—long considered a laggard to Nvidia—has slowly gained ground over the past year

Prior to making GPUs in the mid-2000s, AMD was Intel Corporation’s (INTC) chief rival in the CPU market. However, after a series of strategic moves such as spinning off its foundry business and designing better value CPUs, AMD eventually chipped away at Intel’s lead. 

By 2018, AMD pivoted to cloud computing, and launched its own GPUs for data center and AI workloads. While the company has largely trailed behind Nvidia, the chip designer has positioned itself to take advantage of AI data center demand over the past year.

Last year, AMD landed multiple AI data center deals. Among the notable ones is the partnership with AI heavyweight OpenAI where the latter would purchase and deploy 6 gigawatts’ worth of AMD’s Instinct GPUs over the next few years. 

Similar deals were also secured with Oracle (ORCL) and Microsoft (MSFT) where AMD would supply both with Instinct GPUs and Epyc CPUs to be used for both of the firms’ AI data centers and other AI-related offerings.

Earlier this year, AMD bagged a deal worth $100 billion with Meta (META) where the latter will buy 6 gigawatts’ worth of Instinct GPUs to power its data centers. 

Aside from GPUs, AMD’s momentum is being powered partly by its CPU offerings. Demand for its high-performance Epyc CPUs has risen due to increasing data center demand for x86 chips. Another factor fueling demand is the rise of agentic AI, which can be powered through high-performance CPUs.

AMD’s AI-related efforts have enabled it to improve its profitability and valuation. The company’s stock is up 55% year-to-date. Meanwhile, last year’s revenues rose $34.6 billion, up by 34% year-over-year.

The company’s data center segment saw strong growth too, reaching revenues of $16.6 billion, up by 32% year-over-year. The client and gaming segment delivered $14.5 in sales, representing a 51% year-over-year increase.

Last year, AMD posted a Uniform return on assets (“ROA”) of 20% alongside a Uniform asset growth of 35%. 

AMD currently trades at a Uniform P/E of 40x, a signal that the market is pricing in elevated growth expectations amid intense competition and cyclical demand.

While AMD’s Uniform P/E is elevated relative to corporate averages, it has established itself as a viable alternative to Nvidia, whose highly sought-after chips have led hyperscalers scrambling for other sources of high-performance chips.

In all, AMD’s growing data center presence, product improvements, and exposure to the AI boom position it for continued earnings expansion and attractive long-term returns. 

Best regards,

Joel Litman & Rob Spivey
Chief Investment Officer &
Director of Research
at Valens Research