Five semiconductor heavyweights are racing to capture a server CPU market that Bank of America projects will reach $170 billion by 2030.
Five semiconductor heavyweights are racing to capture a server CPU market that Bank of America projects will reach $170 billion by 2030.

Five semiconductor heavyweights are racing to capture a server CPU market that Bank of America projects will reach $170 billion by 2030.
Advanced Micro Devices Inc., Intel Corp., Arm Holdings Plc, Nvidia Corp. and Qualcomm Inc. are ramping AI server processor efforts as agentic AI fuels demand for inference computing, with the market projected to exceed $170 billion by 2030, according to Bank of America Securities.
"The emergence of agentic AI — systems that can plan and execute tasks autonomously — is creating a step-change in server CPU demand that we haven't seen since the cloud migration of the 2010s," Vivek Arya, semiconductor analyst at Bank of America Securities, said.
Arya raised his server CPU total addressable market estimate to $170 billion from $125 billion, implying a 37% compound annual growth rate from 2025 through 2030. He named AMD the bank's top pick in the space, citing its long-term market position and the upcoming Venice server processor launch, and raised his price target to $560 from $500.
The server CPU market's expansion represents a structural shift in how AI workloads are deployed. While Nvidia dominates AI training with its graphics processing units, the inference phase — where trained models generate answers — increasingly runs on central processing units, opening a new battleground among the five chipmakers.
AMD's Product Offensive Gains Traction
AMD has moved aggressively to capture inference share. Its Ryzen AI Halo developer platform, priced at $3,999, undercuts Nvidia's DGX Spark by $700 while claiming support for up to 200 billion parameter models running locally. AMD says the platform delivers up to 14% better tokens-per-second performance across multiple models and runs both Windows 11 and Linux, whereas the DGX Spark is Linux-only.
The product launch follows a string of bullish analyst calls. Citi upgraded AMD to Buy with a $575 target from $460, and the stock hit a record $558 on Monday, pushing the company's market capitalization above $900 billion for the first time. AMD's trailing price-to-earnings ratio of 169x reflects investor expectations for 91% year-over-year quarterly earnings growth.
Intel and Nvidia Stake Their Claims
Intel, which has posted a 500% one-year gain, is leaning on its Xeon 6+ processors and rackscale AI systems to compete. Wells Fargo set a $110 target on the stock, while Barclays assigned $100. Intel's foundry business, where it is building chips on the 18A node (equivalent to roughly 1.8nm), also positions it as a manufacturing partner for rivals' designs.
Nvidia, meanwhile, is not standing still. The company reported $81.62 billion in quarterly revenue and authorized an $80 billion share buyback. Chief Executive Officer Jensen Huang has positioned the new Vera central processing unit as a potential $200 billion opportunity, indicating that Nvidia views the CPU market as complementary to its GPU dominance rather than a threat.
Arm and Qualcomm are also pushing into the server CPU space. Arm's energy-efficient architecture, already dominant in smartphones, is gaining traction in data centers through partnerships with cloud providers. Qualcomm is using its Snapdragon compute expertise to target AI inference workloads at the edge and in the cloud.
What This Means for Investors
The $170 billion server CPU opportunity is not evenly distributed. AMD and Intel are best positioned in the traditional x86 server market, while Arm-based challengers threaten to erode their share over time. Nvidia's ability to bundle CPUs with its GPUs gives it a systems-level advantage that competitors cannot easily replicate.
AMD shares, trading at 169x trailing earnings, price in aggressive growth from the Venice launch and the MI450 accelerator ramp. Intel, at a fraction of that multiple, offers a turnaround story that depends on foundry execution. For investors, the key swing factors are hyperscaler capital expenditure commitments and the pace at which agentic AI workloads shift from training to inference.
This article is for informational purposes only and does not constitute investment advice.