Global Economic Times

Jensen Huang at CES 2026: "AI to Trigger Storage Revolution and Exploding CPU Demand": Global Economic Times Reporter / Updated : 2026-01-07 07:05:55

LAS VEGAS — Jensen Huang, CEO of NVIDIA, declared that Artificial Intelligence is set to ignite a fundamental revolution in storage technology, driven by a radical departure in how data is processed and remembered.

Speaking at a global press conference at the Fontainebleau Hotel in Las Vegas on January 6 (local time), Huang highlighted the critical role of the Key-Value (KV) Cache—a specialized form of short-term memory that stores context from previously processed data.

"AI utilizes memory in a way that is entirely different from traditional IT industries," Huang explained. "Because of how KV Cache is managed, the demand for high-performance Central Processing Units (CPUs) that oversee these storage processes is exploding."

NVIDIA’s Ascent in the CPU Market
Huang signaled NVIDIA's ambition to dominate the CPU landscape, much as it has with GPUs. In NVIDIA’s "Grace" CPU architecture, 16 Low-Power Double Data Rate (LPDDR) DRAM units are integrated to store essential context and task information, allowing AI to maintain coherent conversations and workflows.

"CPUs will be utilized in countless applications," Huang noted. "It would not be surprising at all if NVIDIA becomes the world’s number one CPU company."

A "Big Spender" for Memory Giants
The CEO emphasized that NVIDIA has become one of the most significant buyers in the global semiconductor market. Beyond High Bandwidth Memory (HBM), which is essential for AI training, NVIDIA is also a massive consumer of Graphics DRAM (GDDR) and LPDDR.

"We are a significant consumer of GDDR," Huang stated, confirming that the company continues to source HBM, GDDR, and LPDDR from the "Big Three" memory suppliers: Samsung Electronics, SK Hynix, and Micron.

The Era of the "AI Factory"
Looking ahead, Huang predicted a global surge in the construction of "AI Factories"—dedicated data centers designed to produce intelligence at scale. This trend, he suggested, creates an incredibly favorable environment for memory manufacturers.

"The world will need many more AI Factories," Huang concluded. "In this era, it is a very good time to be a memory supplier."