Investors are keenly anticipating Nvidia's upcoming fiscal 2027 second-quarter earnings report, expected on August 26. As a pivotal player in the burgeoning artificial intelligence sector, Nvidia's performance and CEO Jensen Huang's insights into market demand are of widespread interest. While the company's ambitious projection of $1 trillion in sales from its Blackwell and Vera Rubin GPUs between 2025 and 2027 remains a primary focus, a less prominent but equally significant area of its business is rapidly gaining traction and warrants close attention from investors.
Nvidia's financial reporting consolidates all GPU revenue under its data center division, which now constitutes the vast majority of the company's earnings. In fiscal year 2026, the data center segment alone generated nearly $194 billion. Current analyst estimates from Visible Alpha project this revenue to surge to approximately $368 billion in fiscal 2027 and further to $531 billion in fiscal 2028. These figures collectively align with the company's ambitious multi-trillion-dollar sales target. Therefore, as the earnings report approaches, investors will be scrutinizing any adjustments, upward or downward, to these fiscal 2027 and 2028 data center revenue forecasts.
Beyond the impressive GPU business, a critical new development in Nvidia's portfolio is its venture into the Central Processing Unit (CPU) market, a segment that, while traditionally associated with older consumer electronics, is experiencing a significant revival thanks to agentic artificial intelligence. Agentic AI focuses on enabling autonomous tasks with minimal human intervention. In this evolving landscape, while GPUs remain essential for processing complex AI computations, CPUs are increasingly recognized for their vital role in executing these tasks, including planning and accessing external data sources.
Last quarter, Nvidia introduced its Vera CPU, specifically engineered for agentic AI applications. CEO Jensen Huang highlighted during the recent earnings call that this new product line opens up a substantial $200 billion total addressable market. He further indicated that Nvidia anticipates generating approximately $20 billion in CPU revenue this year alone, positioning the company as a major, if not leading, force in the CPU domain almost instantly. Huang articulated his vision, stating that just as the world has billions of human users, it will eventually host billions of AI agents, implying immense growth potential for CPU demand.
This optimistic outlook on the CPU market is not exclusive to Huang. Intel's CEO, Lip-Bu Tan, has also observed a significant shift in the ratio of CPUs to GPUs utilized for AI inference, moving from 1:8 to 1:4, with a potential for parity or even dominance in the future. A June research note from Bank of America analysts reinforces this sentiment, forecasting a fivefold expansion of the CPU total addressable market, from $35 billion in 2025 to an astounding $170 billion by 2030. While Bloomberg's projections for the GPU TAM indicate a rise to $486 billion by 2033, translating to approximately $328 billion by 2030—still significantly higher than the CPU TAM—the dynamic nature of AI development suggests that these market shares could rapidly evolve. Therefore, investors should remain attentive to any updated guidance and remarks from Huang regarding the CPU segment, as it promises to be a crucial growth driver for Nvidia and the broader AI industry.
