Demand for artificial intelligence (AI) compute has shifted significantly in 2026. While hyperscalers had focused the majority of their spending on graphics processing units (GPUs) and custom AI accelerator chips over the last three years, they’re now adding tons of central processing units (CPUs), and a handful of companies are seeing the benefits.

Arm Holdings (NASDAQ: ARM) has been one of the biggest beneficiaries so far. While it’s long held a dominant position in mobile device CPUs, it’s made significant progress in data center CPUs over the past few years. That’s a market historically dominated by Intel and (to a lesser extent) AMD. But energy-efficient Arm-based designs could become an extremely valuable resource for power-hungry AI data centers, potentially accelerating Arm’s revenue growth going forward.

Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a “Double Down” signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same “Total Conviction” signal is flashing for a company 1/100th the size of Nvidia. Continue »

The market has taken notice of Arm’s momentum in the area, and the stock has tripled in value since the start of the year. Investors may be wondering if they missed a big opportunity or if the semiconductor stock can triple again from here.

Image source: The Motley Fool.

Can Arm dominate demand for data center CPUs?

At Arm’s investor day back in March, it estimated the total addressable market for data center CPUs will reach $100 billion by 2031, up from $50 billion in 2026. That may be conservative. During its most recent earnings call, AMD updated its server CPU market estimate to $120 billion by 2030.

Driving the growth in CPU demand is agentic AI. CPUs are great at orchestrating a server of GPUs, determining where data should move, and handling machine-to-machine communication needs. If a GPU is specialized labor, a CPU is the manager. As agentic workloads grow, CPU demand increases. Intel CEO Lip-Bu Tan suggested that the ratio of CPUs to GPUs could move from 1:4 currently to 1:1 in the future as more work migrates to AI agents and they handle more complex tasks.

Importantly, all indications are that Arm is seeing tremendous market share improvements in the data center space. Management said Arm’s market share of CPU compute among the top hyperscalers is about 50% in its fourth-quarter letter to shareholders. That’s backed up by statements from Nvidia and Amazon, which use Arm architecture for their CPUs.

Source link
More: https://theglobaltrack.com/
https://corinthiames.com.br/