What Did NVIDIA Say in the Closed-Door Roadshow? What Rumors Were Addressed?

Author: qinbafrank

What did NVIDIA say in the closed-door roadshow? What rumors were addressed? This week, Morgan Stanley organized a non-deal roadshow for NVIDIA in California, attended by Jensen Huang, NVIDIA's CFO Colette Kress, and IR head Toshiya Har, who met intensively with investors. The closed-door meeting mainly aimed to clarify and respond to market concerns regarding product progress, ASIC competition, and sustainable growth (FUD). Here are the core contents and key points of the closed-door meeting:

1. Revenue Growth and Overall Outlook (Most Concerned)

1. Quarterly revenue has reached $100 billion, and the growth rate is still accelerating. This is seen as a strong signal, indicating that demand far exceeds previous expectations.

2. The Vera Rubin product cycle is robust, expected to bring significant orders and revenue increases over the next 12 months, serving as the main driver for short-term growth.

3. Institutions like Morgan Stanley have given optimistic forecasts: approximately $393 billion in revenue for fiscal year 2027 and about $598 billion for fiscal year 2028.

4. CPU business target: striving for $20 billion this fiscal year.

2. Product Roadmap and New Product Progress (Key Rumor Refutation)

1. Rubin Ultra: It is confirmed to ship next year (2027), completely denying rumors of a delay to 2028. The roadmap remains unchanged.

2. Kyber Rack: The current design is a transitional solution, which will be replaced by a better solution to support larger-scale vertical expansion of single racks (larger computing clusters). This is an optimization adjustment, not a delay or issue; the roadmap maintains flexibility to reduce capacity ramp-up risks.

3. Core technologies are progressing as planned:

◦ 800V power supply system.

◦ Optical vertical expansion between racks (optical links/inter-rack optical vertical expansion).

• Overall next-generation product cycle (including the Rubin series) has no significant changes, with Jensen Huang personally clarifying, demonstrating the company's confidence in the progress.

3. ASIC Competition and Market Share

1. Even with some cloud vendors and cutting-edge AI labs heavily investing in custom ASICs (application-specific integrated circuits), NVIDIA still maintains a strong dominant position.

2. Key case: A major client that previously relied mainly on ASICs for cutting-edge model development has significantly increased its usage of NVIDIA GPUs to nearly 50% (market speculation may point to labs like Anthropic). This is seen as a significant validation— even the ASIC camp needs more NVIDIA GPUs.

3. Reason: NVIDIA has the lowest token usage cost, and its full-stack ecosystem (GPU + networking + software + CPU, etc.) forms a strong moat, with cutting-edge AI workloads still highly dependent on its platform.

4. AI labs currently account for about 20% of AI computing power demand, with significant growth in their demand.

4. Customer Diversification and New Growth Engines

1. Customer structure has shifted from a past heavy reliance on a few large hyperscalers to diversification: AI labs, enterprise clients, emerging cloud service providers (neocloud), and Sovereign AI projects (data security, industrial autonomy, geopolitical factors driving increased importance).

2. New growth points (enterprise, Sovereign AI, industrial sector clients) may grow faster than traditional hyperscalers, reducing concentration risk.

5. Supply Chain and Supply Constraints

1. The supply shortage of memory (HBM, etc.) is expected to last for several years. The growth rate of AI token demand is faster than the expansion of storage/memory capacity, leading to an industry-wide shortage (affecting data center components, GPUs, and AI operations).

Jensen Huang clearly stated that the memory shortage is a long-term phenomenon.

2. Overall supply remains the main bottleneck, but NVIDIA is responding through a diversified supply chain and planning.

6. Other Management Views and Capital Returns

1. Strong free cash flow (FCF) supports shareholder returns: more than 50% of future cash flow may be used for buybacks and other shareholder returns.

2. The moat of the full AI infrastructure platform (hardware + networking + software) continues to strengthen.

Morgan Stanley Summary:

Market concerns about product delays, competition, and sustainable growth have been exaggerated, and NVIDIA is "firing on all cylinders."

Morgan Stanley maintains a strong bullish stance and raises the target price to $288 (indicating significant upside potential).

Disclaimer: This content is provided for general branding and informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online events, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets or to use any services. Crypto assets are highly volatile and may result in loss. WEEX services and online events may not be available in all regions and are subject to applicable laws, regulations, and eligibility requirements. You are responsible for ensuring that your use of WEEX services complies with local laws and for carefully assessing the risks before participating in any crypto-related activities.