Arm Holdings positioned to lead AI inference chip market as Nvidia, AMD, Broadcom, and Intel compete for dominance

Yahoo Finance AI · May 23, 2026

AI Summary

•Deloitte projects inference workloads will account for two-thirds of AI computing power in 2026, up from 50% in 2025, with the inference-focused AI chip market potentially reaching $50 billion this year. McKinsey estimates AI inference workloads in data centers could jump from almost 21 gigawatts in 2025 to 93 GW in 2030, at a compound annual growth rate (CAGR) of 35%.
•Arm Holdings licenses its energy-efficient chip architecture to major companies including Nvidia (which uses Arm's design for its Grace server CPU and Vera CPU), Google, Amazon, Broadcom, MediaTek, Qualcomm, and Apple for running AI inference across data centers, smartphones, and laptops. Arm's royalty rate for its AI-focused Armv9 architecture is nearly double that of the previous generation Armv8 architecture.
•Arm expects royalty revenue to increase at a CAGR of 20% between fiscal 2026 and 2031, compared to 14% annual growth over the past five years. The company estimates its own CPU could generate $15 billion in annual revenue in fiscal 2031 and projects overall revenue of $25 billion in fiscal 2031, a potential increase of more than 5x compared to the $4.7 billion revenue in the trailing twelve months, with non-GAAP earnings projected to exceed $9.00 per share in fiscal 2031 versus $1.77 in fiscal 2026—a 39% CAGR.

Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.