Nvidia CEO Says Demand for Computing Resources Is 'Skyrocketing'

Nvidia CEO Jensen Huang announced the full production of the company’s next-generation AI chips, the Vera Rubin platform, at CES 2026. These chips are designed to deliver five times the artificial-intelligence computing power of previous generations. The platform is set for deployment later this year by major cloud providers, including Microsoft, AWS, and Google Cloud. The new architecture also includes a new layer of storage technology aimed at improving chatbot performance.

The launch of the Vera Rubin platform comes amid significant demand for AI infrastructure, driven by the expanding use of AI in enterprise applications. Huang emphasized that the new chips will allow companies to run large AI models more efficiently while reducing costs per token by up to 10 times compared to the previous Blackwell architecture. This performance gain is expected to help Nvidia maintain its leadership in the AI chip market, despite growing competition from AMD and in-house solutions from cloud providers like Google and Meta.

Rising memory costs are pushing GPU prices higher for both Nvidia and AMD, as demand for AI data centers intensifies. A Kbench report states that prices of GDDR7 and GDDR6 memory have surged several hundred percent in recent months. This cost pressure is leading both companies to implement phased price increases across their GPU portfolios in early 2026. The GeForce RTX 5090 could reach $5,000 by mid-2026, while mid-range GPUs like the RTX 5070 may see production cuts of 30% to 40%.

Why Is This Happening Now?

The surge in demand for computing resources is driven primarily by the rapid expansion of AI workloads, which require high-performance hardware for training and inference tasks. This demand has shifted the supply of GPUs away from traditional gaming customers toward data centers and AI enterprises. As a result, consumer GPU prices remain elevated, and supply is limited.

Nvidia is also expanding its infrastructure offerings with new networking and storage technologies to support the Vera Rubin platform. These include co-packaged optics and a new BlueField-4 DPU, which help improve interconnectivity and power efficiency in large-scale AI deployments. The company’s strategy is to provide a complete ecosystem for AI workloads, reducing dependency on third-party components.

How Are Markets Reacting?

The stock market has responded positively to Nvidia’s announcements. Investors remain bullish on AI-driven growth, particularly as the company secures major licensing deals like the $20 billion agreement with Groq for access to inference processing units. Wedbush analysts note that demand for AI systems from hyperscalers will remain strong in 2026, despite rising memory costs that could lead to demand destruction in consumer devices.

However, concerns about margin sustainability are growing. Multiple analysts question whether Nvidia can maintain its historically high profit margins amid rising competition and pricing pressure. Wedbush analysts also highlighted the risk of margin compression due to large licensing and acquisition costs.

Nvidia’s recent earnings report showed strong results, with revenue of $57.01 billion in the latest quarter, surpassing analyst expectations. The company reported a net margin of 53.01% and a return on equity of 99.24%. These numbers suggest continued profitability, but investors remain cautious about long-term margin pressures.

What Are Analysts Watching Next?

Analysts are closely monitoring the impact of rising memory costs on consumer GPU sales. While demand for AI remains strong, there are concerns that elevated GPU prices could reduce demand for gaming and personal computing hardware. This could affect the broader PC market, which relies on GPUs from companies like Nvidia and AMD.

Regulatory risks are also a concern. Reports of alleged smuggling of export-controlled Nvidia chips into China have raised the possibility of increased regulatory scrutiny and reputational damage. This could impact long-term sales in the region, where demand for AI and data center hardware remains high.

Looking ahead, the Vera Rubin platform is expected to drive further revenue growth for Nvidia in 2026. With major cloud providers already planning to deploy the new chips, the company is well-positioned to continue its dominance in the AI hardware market. However, the company must also address growing competition from AMD and in-house solutions developed by cloud providers and AI labs.

The success of the Vera Rubin platform will likely depend on the broader adoption of Nvidia’s proprietary technologies and its ability to maintain cost advantages over alternative solutions. As AI infrastructure spending accelerates, the pressure to innovate and deliver value will only increase.