New Report Finds Semiconductors Account for 95% of an AI Data Server Rack’s Value, Encompassing the Full Stack of Chip Technologies

A joint study by the Semiconductor Industry Association (SIA) and Deloitte reveals that semiconductor technologies now represent 95% of the content value within state-of-the-art AI data server racks. The report projects that annual revenue from chips deployed in AI data centers could surge to $1.2 trillion by 2028, marking a nearly tenfold increase over a four-year period. This massive growth underscores the critical role of the full semiconductor stack—including logic, memory, and analog components—as the foundational engine for the global AI infrastructure market.

The report, titled “Powering AI: The Semiconductor Ecosystem at the Foundation of Data Centers,” provides a detailed virtual teardown of modern AI infrastructure to illustrate the dominance of chip technology. It finds that semiconductors account for more than 95% of a leading AI server rack’s value and represent over 50% of the total capital expenditures (CapEx) required to build and operate an AI data center. This high concentration of value highlights how AI development is fundamentally dependent on a broad spectrum of hardware, ranging from high-end logic and memory to essential analog and foundational chips.

Financial projections within the study suggest a transformative shift in the market, with AI-related chip revenue expected to hit $1.2 trillion by 2028. This figure is particularly significant as it would surpass the total global semiconductor sales recorded in 2025 across all end-use categories by more than 50%. The SIA notes that this growth represents a nearly tenfold increase in just four years, positioning AI data centers as the primary engine for the semiconductor industry's future expansion and revenue generation.

From a policy perspective, the report arrives as the U.S. government advances initiatives like the Trump Administration’s Pax Silica Initiative and the AI Exports Program to maintain domestic leadership. John Neuffer, SIA president and CEO, emphasized that the future of AI hinges on the full array of chip technologies and called for government policies that ensure American chips remain at the center of global innovation. As competition intensifies, the report suggests that the industry must navigate rapid technological changes and evolving market trends to maintain its competitive edge in the global AI landscape.