5 trillion revenue forecast is too conservative! Nvidia CFO says "it will definitely be higher," Jensen Huang says demand from Chinese customers is strong.

5 trillion revenue forecast is too conservative! Nvidia CFO says "it will definitely be higher," Jensen Huang says demand from Chinese customers is strong.

Nvidia executives have repeatedly given optimistic signals, expressing confidence in the sales prospects of the Blackwell and next-generation Rubin chips.

On Tuesday, the 6th Eastern Time, Nvidia CFO Colette Kress said at a JPMorgan-hosted event that due to strong demand, Nvidia is now more optimistic about its data center business. By the end of 2026, the expected revenue from Nvidia’s data center chips will “definitely” exceed the $500 billion forecast given last October.

The $500 billion mentioned by Kress refers to what Jensen Huang stated at the GTC conference over two months ago: that by the end of 2026, Nvidia’s existing and future data center chips will generate about $500 billion in revenue. Goldman Sachs pointed out at the end of October that this forecast was about 12% higher than the market consensus at the time.

Kress’ comments continued the optimism expressed by Nvidia CEO Jensen Huang at this week’s Consumer Electronics Show (CES). On Monday, Nvidia officially announced the release of the Rubin chip. In his CES keynote, Huang revealed that the new generation Vera Rubin platform is now in full-scale production and the inference cost has been reduced to one tenth of the Blackwell platform.

Huang also emphasized in a media interview on Tuesday that Nvidia’s new chips deliver 10 times the performance of the previous generation, and said that Chinese customer demand is strong.

Despite positive signals from both Kress and Huang, Nvidia’s stock price has not reversed its decline. On Tuesday, Nvidia’s shares rose more than 2% in early trading but then continued to fall, turning slightly negative at midday, dropping over 0.4% at their lowest point of the day, and may close down for a second consecutive session. Nvidia has risen about 39% in 2025, still a major driver of US stock gains, but far less than the more than 170% increase in 2024, mainly due to concerns about an AI bubble and competition from Google’s TPU.

Revenue Forecast Continues to Rise

In her remarks on Tuesday, Kress clearly stated that since giving the $500 billion forecast in October 2025, Nvidia’s customer interest has continued to increase. Speaking about the data center chip revenue forecast, she said, “The $500 billion (forecast) figure will definitely be even higher.”

Kress pointed out that Nvidia’s positive outlook for AI applications stems not only from AI demand itself, but also from increasing enterprise data processing needs, which are driving growth in next-generation computing demand. This will help overall investment reach trillions of dollars by the end of 2030.

This statement echoes the prediction made by Jensen Huang at the GTC conference on October 28, 2025. At that time, Huang revealed that the company has “visibility” on achieving a cumulative $500 billion in data center business revenue between 2025 and 2026. This forecast covers both Blackwell and next-generation Rubin architecture products. According to Goldman Sachs analysis, this target is 12% higher than the $447 billion consensus reflected in Visible Alpha Consensus Data, and also 10% higher than Goldman Sachs’ own forecast of $453 billion.

In early December 2025, Kress stated that the $500 billion in bookings did not include any work Nvidia was doing on the next phase of the OpenAI agreement, and revealed that the previously announced $100 billion investment agreement with OpenAI had not yet been finalized. She said at the time there was “definitely an opportunity to receive more orders on top of the announced $500 billion.”

At Tuesday’s event, Kress was asked about the role the Chinese market might play in Nvidia’s revenue. Kress said the U.S. government is working to approve license applications; Nvidia has received orders from clients, but the specific outcome is still unclear.

Rubin Platform Goes Full-scale Production

On Monday, Jensen Huang announced at CES in Las Vegas that the new generation Vera Rubin AI platform has gone into full-scale production. He stated that through the integrated design of six new chips, the platform achieves significant leaps in inference cost and training efficiency and will begin delivering to initial customers in the second half of 2026.

Huang described the Rubin GPU as “a giant monster,” and elaborated on the design logic: “AI inference costs must drop tenfold every year, while the number of tokens generated by AI ‘thinking’ grows fivefold annually.” He emphasized that this leap in performance beyond Moore’s Law expectations is the result of “extreme collaborative design”—thorough reengineering from CPU, GPU, network chips to cooling systems.

Reportedly, the Rubin GPU achieves 50 PFLOPS inference performance at NVFP4 precision, five times that of Blackwell; training performance is 35 PFLOPS, a 3.5-fold increase over the previous generation. Inference token generation cost can drop as much as one-tenth of the Blackwell platform. Each GPU package contains eight sets of HBM4 memory with bandwidth reaching 22 TB/s. Microsoft’s next-generation AI superfactory will deploy hundreds of thousands of Vera Rubin chips.

To solve the AI “memory” bottleneck, Nvidia has built an inference context memory storage platform based on BlueField-4 DPU, adding 16TB of high-speed shared memory to each GPU on top of the existing 1TB memory, connected via 200Gb/s bandwidth, solving the “VRAM wall” problem for long texts.

In terms of energy efficiency, the Rubin NVL72 rack achieves 100% liquid cooling, supporting inlet water temperatures of 45°C, meaning data centers can dissipate heat without high-energy chilled water units. Huang stated that this will save 6% of global data center electricity.

Jensen Huang’s Speech Led to Storage Stocks Soaring, Data Center Cooling Stocks Plunging

Jensen Huang’s speech at CES had a significant impact on related industry stock prices. In his address, he highlighted the demand for memory and storage in AI systems: “This is a market that has never existed before and is likely to become the world’s largest storage market, essentially bearing the working memory of global AI.”

Boosted by these comments, storage chip giant SanDisk became the best-performing S&P 500 constituent stock on Tuesday, surging 25% intraday and poised to close up over 20%, marking its largest single-day gain since February 18, 2025.

SanDisk rose over 40% in the first three trading days this year, and has soared about 1,050% from its low in April last year. Storage device manufacturers Western Digital and Seagate Technology both logged double-digit gains on Tuesday.

Data center cooling system manufacturers saw sharp declines. Johnson Controls’ shares fell as much as 11% on Tuesday, the largest drop since 2022. Modine Manufacturing plunged 21% intraday before paring losses by half, while Trane Technologies and Carrier Global also declined.

The declines came after Huang stated that Rubin chips can be cooled with warm water without chilled water units. At CES, Huang said server racks equipped with the new Rubin chips can use cooling temperatures that don’t require chilled water systems, with airflow needs similar to those for racks with Blackwell chips.

Industry research notes that chilled water systems are “primary” equipment provided by companies like Trane and Johnson Controls to data centers.

Baird analyst Timothy Wojs wrote in a report that Huang’s remarks “raised some questions and concerns about the long-term positioning of cooling equipment in data centers, especially as liquid cooling technology grows increasingly prevalent.”

Barclays analyst Julian Mitchell pointed out, “Given Nvidia’s dominant position in the entire AI ecosystem, its comments may seem exaggerated at first, but they should not be underestimated.”

Risk Warning and DisclaimerThe market has risks, and investment must be cautious. This article does not constitute personal investment advice, nor does it take into account the individual investment objectives, financial situation, or needs of specific users. Users should consider whether any opinions, views, or conclusions in this article are appropriate for their particular situation. All investment decisions are at your own risk.