NVIDIA responds to rumors of delayed release of next-generation AI server systems: Product roadmap unchanged

The rumor of Nvidia’s next-generation rack-level AI server system Kyber NVL144 being delayed caused a dramatic market reaction.

On Monday, July 6 Eastern Time, Nvidia responded to the rumor by saying its product roadmap has not changed, denying any impact on core progress, but market sentiment was already clearly pressured on the same day. After the rumor broke the previous Sunday, multiple AI hardware supply chain stocks in Japan, Korea, and Taiwan plunged on Monday, highlighting investors’ extreme sensitivity to any “marginal disturbance” in the AI capital expenditure cycle.

Before this volatility, AI-related assets had experienced years of gains with high valuations and expectations, so any unverified supply chain risk signal could be magnified in interpretation.

The “delay controversy” regarding Kyber NVL144 has not yet been officially confirmed, but it has fully exposed a key feature of current AI trading: expectations are highly front-loaded, and actual delivery has become the new anchor. Within this framework, Nvidia’s response may stabilize the narrative in the short term, but the sharp fluctuations in the supply chain indicate that the market has begun to reprice the execution risk of the “next-generation computing power cycle.”

A Leak Triggers Chain Reaction: Kyber NVL144 Delay Uncertainty

According to SemiAnalysis’s analysis published on social media, Nvidia’s code Kyber NVL144 AI server rack system has encountered a technology bottleneck in the high-end PCB (printed circuit board) manufacturing process, causing the project’s delivery time to be possibly delayed by more than 12 months, even pushed to around 2028.

The agency also mentioned that this architecture was one of Nvidia’s key upgrade paths shown after the previous GTC event, originally regarded as an important node in its AI computing power “scale-out” roadmap.

However, according to media reports early in the U.S. stock trading session on Monday the 6th, a Nvidia spokesperson stated by email: “Our road map is intact.”

Nvidia did not disclose the progress details of the Kyber NVL144 project, nor confirm any delays.

Market participants pointed out that even unofficial, unverified supply chain rumors can amplify sentiment during the high-investment cycle for AI infrastructure.

Asian Computing Power Supply Chain Hit With Sell-Off: PCB and Packaging Most Affected

Impacted by the SemiAnalysis rumor, the Asian AI hardware supply chain was generally pressured in trading on Monday this week, especially among PCB and high-end materials sectors.

Japan’s PCB manufacturer Ibiden fell about 10%, regarded as a core supplier to Nvidia. Meanwhile, Hong Kong-listed Kingboard Laminates Holdings plunged about 18%, Taiwan-listed Elite Material Co. closed down about 10%, and Samsung Electro-Mechanics, a component unit under Samsung in Korea, dropped over 10% intraday.

Across the broader sector, MSCI-related semiconductor and tech indices have seen continual corrections recently; some institutional statistics show that cumulative declines have approached 10% in two weeks, indicating the crowding of AI trades is being repriced.

Market commentary generally believes this decline was not driven by a single event, but was a typical reaction under a “highly sensitive structure.”

Market Interpretation: From “Growth Narrative” to “Delivery and Rhythm Risk”

Regarding the Kyber NVL144 rumor, some institutions believe that even if delays occur, it does not necessarily mean that the AI capital expenditure cycle has peaked.

Gary Tan, portfolio manager at Allspring Global Investments, noted that the delay “more likely means that Nvidia’s most aggressive next-generation system architecture will need more time for deployment,” but does not necessarily weaken overall AI infrastructure demand.

However, he also emphasized that the market is now highly sensitive to AI trading, and any uncertainty relating to the supply chain or delivery rhythm could trigger temporary profit-taking and sector rotation.

Korean broker NH Investment & Securities stated that this sell-off reflects “widespread weakness in regional tech sectors”; the Kyber NVL144-related rumor, combined with other AI supply-side concerns, is magnifying investor pricing of uncertainty for next-generation computing power platforms.

In addition, the market is watching several key catalysts this week, including Samsung Electronics’ earnings, the US IPO of memory giant SK Hynix, and industry guidance—all of which could further affect sentiment in the AI chain.

AI Trading Enters “Validation Period”: From Expectation-Driven to Real Constraints

Over the past two years, the core driver of the AI trade has been persistent upgrades in computing power demand and supply chain expansion expectations. At present, the market is gradually entering an “execution validation period”:

On one hand, hyperscale cloud providers’ capital expenditures remain high, and there’s no clear inflection point in demand; on the other hand, bottlenecks in complex rack systems and advanced PCB and packaging processes mean delivery rhythm itself is becoming a new uncertain variable.

Therefore, even though Nvidia stresses that “the roadmap is unchanged,” market reactions are focused more on “rhythm risk” rather than a “collapse in demand.” This also explains why price fluctuations of related supply chain stocks have been amplified—under the backdrop of high valuations, the market is extremely sensitive to any “delay narrative.”

Risk Warning and DisclaimerThe market involves risks; investing requires caution. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial conditions, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular situation. Invest accordingly at your own risk.