NVIDIA Blackwell AI Servers Face Component Shortage

NVIDIA's upcoming Blackwell AI product line is expected to face supply chain obstacles, impacting its market introduction. The Blackwell series aims to transform the industry by integrating advanced technologies, but this innovation is introducing new supply chain complexities.

A recent Bank of America report highlights that NVIDIA’s Blackwell servers are experiencing shortages of crucial industrial components, which are affecting delivery schedules from key Taiwanese suppliers, including Foxconn and Wistron. The main challenge involves the GB200 Blackwell AI servers, which use liquid cooling technology. Shortages of essential components for this cooling system have resulted in significant order backlogs.

This issue is reminiscent of the supply challenges faced by NVIDIA’s previous Hopper generation, although the Blackwell series has not yet launched. The report suggests that if supply problems persist with the Blackwell architecture, NVIDIA might promote its Hopper products, such as the H200 AI GPU, to alleviate the impact. While there is no confirmed delay for Blackwell, it is evident that supply limitations will affect its release.

Despite these challenges, NVIDIA has a track record of handling supply chain issues for new product launches and has prepared contingency plans. The Hopper chips, which continue to drive significant revenue for NVIDIA's AI segment, may serve as a temporary solution. The Blackwell AI products are projected to contribute substantially to NVIDIA’s revenue by 2025, with mass production expected to begin in the latter half of 2024.

Given that Blackwell is anticipated to be one of NVIDIA’s most successful and high-value products, the company is well-positioned to manage these supply chain disruptions effectively. Initially, Blackwell may be available exclusively to select partners such as Meta and Microsoft.