Bank of America raises its 2030 server CPU market size forecast to US$240 billion, expects AMD (AMD.US) to be a major beneficiary.
In its latest semiconductor industry research report, Bank of America points out that as AI applications accelerate their evolution from traditional large model training and inference toward agentic AI, the importance of central processing units in data centers is rising significantly.
Bank of America Corp noted in its latest semiconductor industry research report released on November 12 that as AI applications accelerate their evolution from traditional large-model training and inference toward agentic AI, the importance of central processing units (CPUs) in data centers is rising significantly. The bank raised its forecast for the 2030 global server CPU total addressable market from approximately US$210 billion previously to more than US$240 billion, and believes AMD (AMD.US), with its leading product portfolio and supply capabilities, is poised to become a major beneficiary of this growth trend.
Bank of America Corp analysts Vivek Arya and his team expect the global server CPU market to grow from approximately US$35 billion in 2025 to US$64.1 billion in 2026, and to reach US$240.2 billion by 2030, nearly a sixfold increase from 2025, with a compound annual growth rate of 39.2% from 2026 to 2030, up from the previous forecast of approximately 36%.
Despite concerns that AI capital expenditure may slow after 2028, Bank of America Corp noted that judging from GPU spot rental prices and memory spot prices, AI computing demand is currently still at historic highs, while agentic AI penetration remains low, meaning related infrastructure demand is still in the early stages of growth.
Bank of America Corp believes the core question for future growth in the server CPU market is whether AI demand can continue to drive simultaneous increases in both CPU unit shipments and average selling prices. The bank is optimistic on this front, expecting global server CPU shipments to rise from 38.6 million units in 2026 to 86.2 million units in 2030, a compound annual growth rate of 22.3%; over the same period, average selling prices are expected to rise from US$1,660 to US$2,787, a compound annual growth rate of 13.8%.
By specific application area, Bank of America Corp expects that by 2030, the CPU market for traditional enterprise servers and cloud computing infrastructure will reach approximately US$30 billion, while the CPU market for AI computing cluster management and coordinating GPU operations will exceed US$105 billion, and the CPU market specifically for agentic AI task scheduling and execution will also reach approximately US$105 billion.
This means AI-related applications will contribute approximately US$210 billion of server CPU demand, accounting for approximately 87% of the overall market. Among this, the agentic AI-related CPU market is expected to grow from approximately US$6.8 billion in 2026 to US$105 billion in 2030, a compound annual growth rate of nearly 99%.
Unlike traditional chat Siasun Robot&Automation, agentic AI can not only answer user questions but also autonomously plan tasks, invoke external tools, execute multi-step operations, and continuously process and manage information. These functions require CPUs to handle substantial task scheduling, memory management, and data processing work, thereby further enhancing the value of CPUs in AI infrastructure.
Bank of America Corp noted that CPUs in current AI servers are primarily responsible for coordinating GPU operations, managing computing clusters, and transmitting data. However, as agentic AI applications become increasingly complex, standalone CPU servers and racks dedicated to executing related tasks will become a new growth area. For example, NVIDIA Corporation's (NVDA.US) Vera CPU dedicated rack is a reflection of this trend.
According to Bank of America Corp's estimates, under the current resource allocation model for agentic AI applications, every 1 billion users may generate additional demand for approximately 2.7 million CPUs. However, as agentic AI gradually becomes an important tool for individuals and enterprises in their daily work, the computing resources required per user will increase significantly, and future CPU demand per 1 billion users could rise to approximately 21 million units.
The bank expects that by 2030, the global agentic AI user base could reach 1 billion to 2 billion people, and annual CPU shipments specifically for agentic AI will increase from approximately 2.5 million units in 2026 to 28.7 million units in 2030, a more than tenfold increase.
It is worth noting that Bank of America Corp emphasized that growth in CPU demand does not mean the importance of GPUs will decline. On the contrary, the development of agentic AI will further expand the overall market size of AI computing systems, with CPUs and GPUs playing complementary roles in different types of computing tasks.
The bank expects that by 2030, the global data center systems market will reach approximately US$2.8 trillion, of which the AI data center systems market will be approximately US$2.4 trillion and the AI accelerator market will be approximately US$1.6 trillion. Server CPUs' share of the total data center systems market value is expected to rise from less than 7% in 2024-2025 to approximately 9%.
On individual stocks, Bank of America Corp continues to list AMD as its top pick in the server CPU space, maintaining its "Buy" rating and raising its price target from US$650 to US$700, based on a valuation of 28x 2028 expected earnings per share, up from 26x previously.
Bank of America Corp believes AMD has a broad server CPU product portfolio and leading advantages in both high operating frequencies and high core counts. The former helps improve the operating efficiency of AI computing clusters, while the latter helps handle more agentic AI tasks simultaneously.
Specifically, AMD's fifth-generation EPYC Turin server processor has up to 192 cores, while the next-generation Venice processor based on the Zen 6 architecture and 2nm process will have up to 256 cores and 512 threads. Bank of America Corp believes AMD's advantages in product performance, advanced processes, and supply assurance are expected to help it further expand server CPU market share.
The bank expects AMD's server CPU sales to grow from US$18.5 billion in 2026 to US$84.4 billion in 2030, a compound annual growth rate of 46.2%, with its share of global server CPU market sales rising from approximately 29% to 35%.
By contrast, Bank of America Corp holds a relatively cautious view on Intel Corporation's (INTC.US) near-term prospects. Although the bank believes Intel Corporation's foundry business still has potential growth opportunities, its CPU supply outlook for the fourth quarter of this year and 2027 still lacks sufficient certainty. Therefore, Bank of America Corp maintains a "Neutral" rating on Intel Corporation and lowered its price target from US$100 to US$80.
Bank of America Corp expects Intel Corporation's share of global server CPU market sales to decline from approximately 34% in 2026 to 22% in 2030. However, benefiting from overall market expansion and rising average selling prices, Intel Corporation's server CPU sales are still expected to grow from US$21.9 billion in 2026 to US$53.1 billion in 2030, a compound annual growth rate of approximately 25%.
Meanwhile, Arm-based server CPUs are expected to be the fastest-growing product category by market share. Bank of America Corp expects that as NVIDIA Corporation's Vera processor, Arm's (ARM.US) self-developed AGI CPU, Qualcomm's (QCOM.US) server CPUs, and large cloud service providers' in-house chips gradually expand deployment, Arm-based processors' share of global server CPU market sales will rise from 37% in 2026 to 43% in 2030, corresponding to a market size of approximately US$102.8 billion.
However, Bank of America Corp believes that the value created by growth in the Arm-based server CPU market will be captured more by technology licensing customers such as NVIDIA Corporation and large cloud service providers, while Arm's own server CPU products are not expected to achieve shipments of meaningful scale until 2028-2030. The outlook for Qualcomm's server CPU business is similarly uncertain, mainly because it faces multiple competitors among key customers.
Regarding NVIDIA Corporation, Bank of America Corp believes that with its overall competitive advantage in AI computing systems, the company is accelerating the expansion of Vera CPU deployment and is poised to become an important player in the server CPU market. The bank expects NVIDIA Corporation's server CPU sales to grow from US$20.4 billion in 2026 to US$60.6 billion in 2030.
However, given NVIDIA Corporation's recent significant volatility, Bank of America Corp lowered its price target from US$350 to US$280 and cut its valuation multiple from 22x 2027 expected earnings per share to 18x.
The latest industry data also shows that the competitive landscape of the server CPU market is changing. According to Mercury Research data cited by Bank of America Corp, in the third quarter of 2026, global server CPU shipments grew 31% year-over-year and 6% quarter-over-quarter, significantly outperforming the PC CPU market.
Among this, Intel Corporation's server CPU sales grew 14% quarter-over-quarter to US$5.9 billion, slower than AMD's 18% and the estimated 28% for Arm-based processors. Affected by wafer and packaging substrate supply constraints, Intel Corporation's server CPU sales share fell 1.9 percentage points quarter-over-quarter to 34.3%, while AMD and Arm-based processors reached shares of 28.8% and 36.9%, respectively, both near historic highs.
Bank of America Corp believes the development of agentic AI is opening a new growth cycle for the server CPU market. As AI applications gradually shift from pure model training and content generation to agents capable of autonomously executing complex tasks, CPU shipments, average selling prices, and their value in data center systems are all expected to continue rising. Under this trend, chipmakers with leading product performance, ample supply capabilities, and well-developed technology ecosystems will have greater competitive advantages, and AMD remains the bank's most favored server CPU investment target.
Related Articles

New stock news | Hearing passed! Kunlun New Materials' half-year results exceed the whole of last year, and the world's third-largest independent electrolyte manufacturer is about to list on the Hong Kong stock market.

BAMA TEA (06980) responds to suspension of military procurement qualification: it is a qualification adjustment in a specific field and has no impact on the group's public-facing sales, operations, and core business.

YUM CHINA (09987) repurchased approximately 111,500 shares for approximately US$5.5999 million on November 20.
New stock news | Hearing passed! Kunlun New Materials' half-year results exceed the whole of last year, and the world's third-largest independent electrolyte manufacturer is about to list on the Hong Kong stock market.

BAMA TEA (06980) responds to suspension of military procurement qualification: it is a qualification adjustment in a specific field and has no impact on the group's public-facing sales, operations, and core business.

YUM CHINA (09987) repurchased approximately 111,500 shares for approximately US$5.5999 million on November 20.






