AI Infrastructure Boom Drives Demand for Power, Fiber, and Storage Beyond Chips

Key points

• Nvidia stock reached a new all-time high, signaling continued strength in the AI chip sector.
• AT&T and Corning signed a multiyear deal worth over $3 billion for optical fiber, highlighting the physical demands of AI expansion.
• AI data centers require more than 10 times as much optical fiber as traditional facilities, according to Corning.
• Vicor Corp. is positioned to benefit from the need for efficient power delivery to AI processors, which require high current at low voltage.
• MaxLinear and Corning are key suppliers for the networking and fiber needed to connect AI clusters, with MaxLinear supplying silicon and Corning supplying glass.
• Everpure (formerly Pure Storage) is seen as a beneficiary of growing demand for fast, reliable data access for AI applications.

Background

Recent developments in the AI chip sector include Amazon's $60 billion deal with Qualcomm for custom AI chips, Huawei's launch of the Atlas 950 SuperPoD to compete with Nvidia, and Nvidia's production of Groq 3 LPX accelerators for AI inference. These events underscore the intense competition and rapid evolution in AI hardware, with Nvidia maintaining its dominant position in AI training chips.

How outlets are covering it

Yahoo Finance highlights Nvidia's record high as a catalyst for watching other AI chip stocks, while InvestorPlace focuses on the broader AI infrastructure race, emphasizing companies that provide power, fiber, and storage. InvestorPlace notes that the AI economy creates physical demands, such as the need for more optical fiber and efficient power delivery, which are not directly related to chip design but are critical for AI deployment. The two sources differ in their focus, with Yahoo Finance centering on chip stocks and InvestorPlace on the physical infrastructure supporting AI.

Why it matters

The AI boom is driving demand not only for chips but also for the physical infrastructure that supports them, including power, fiber, and storage. This shift highlights the importance of companies that provide these essential components, as AI adoption expands from training models to inference and everyday use. Investors are increasingly looking beyond Nvidia to find opportunities in the broader AI ecosystem, recognizing that the success of AI depends on more than just processing power.

What to watch

Investors will likely continue to monitor the performance of companies involved in AI infrastructure, such as Vicor, MaxLinear, Corning, and Everpure, as AI adoption grows. The success of these companies will depend on their ability to meet the increasing demands for power, fiber, and storage, as well as their valuation and market position. The outcome of the AT&T-Corning deal and the performance of these stocks will provide insights into the health of the AI infrastructure market.