CMB International has released a research report stating that NVIDIA's first-quarter results, ending in April, once again exceeded expectations, with the company also raising its guidance for the second quarter ending in July. However, the more significant insight is that the economic benefits of AI infrastructure are becoming increasingly sustainable. Coupled with visibility for the $1 trillion Blackwell/Rubin platform through 2027, new monetization opportunities from the Vera CPU, and a larger-scale capital return plan, the first-quarter results have alleviated market concerns that fiscal year 2027 might represent only a one-time peak in demand. They also reinforce NVIDIA's transition from a GPU supplier to a comprehensive AI factory platform.

The report indicates that the new business segmentation shows NVIDIA is no longer solely a story about capital expenditure from hyperscale cloud providers. First-quarter revenue from hyperscale cloud customers reached $38 billion, a sequential increase of 12%, while revenue from AI Cloud, Industrial, and Enterprise (ACIE) reached $37 billion, a sequential increase of 31%. This reflects faster sequential growth in AI factories from non-hyperscale cloud customers. The market tends to debate NVIDIA through the lens of hyperscale cloud providers' capital expenditure intensity, ASIC substitution risks, and customer concentration. However, the ACIE segment changes this discussion framework. The firm believes that while hyperscale cloud providers have proven the viability of the architecture, ACIE could represent a broader and higher-quality expansion market.

CMB International stated that the visibility for Blackwell/Rubin and the Vera CPU extend NVIDIA's growth narrative from GPU demand to the control layer of the AI factory platform. The firm views Vera as NVIDIA's move into the CPU-side control layer for AI agents, where the CPU manages orchestration, tool usage, memory, and other functions, while the GPU handles inference. Overall, Blackwell/Rubin provides multi-year platform visibility, while Vera expands NVIDIA's monetization capabilities from GPUs to broader AI factory infrastructure.

The firm also maintains a positive outlook on domestic AI beneficiaries under the global expansion of AI infrastructure capital expenditure. These include leading optical module manufacturer InnoLight Technology, PCB manufacturer Shengyi Technology, and data center power supply manufacturer Innoscience.