Nvidia CEO Jensen Huang believes the company’s explosive growth is far from over. Speaking at the Goldman Sachs Communacopia + Technology conference, Huang reiterated that Nvidia could increase its revenue by roughly 70% next year, arguing that the company’s deep integration across the AI ecosystem gives it an unusually clear view of where the market is heading.
If Nvidia reaches the roughly $400 billion in annual revenue analysts currently expect for its fiscal year, a 70% increase would push next year’s revenue toward approximately $680 billion.
Nvidia Is Selling Far More Than GPUs
Huang’s argument is that Nvidia should no longer be viewed simply as a semiconductor company selling individual GPUs.
The scale of modern AI infrastructure has transformed Nvidia’s products into complete computing systems. Huang pointed to systems combining dozens of Grace CPUs and Blackwell GPUs, with one such system currently seeing sales grow about 27% month over month.
He also highlighted the dramatic increase in the value and complexity of Nvidia’s hardware. A single AI computing system can now cost millions of dollars and contain millions of interconnected components, creating a much larger revenue opportunity than the company’s traditional gaming GPU business.
That shift is central to Nvidia’s growth story: as AI models become more sophisticated, customers are buying increasingly large and integrated computing platforms rather than standalone chips.
Huang Says Nvidia Can “See the Future” of AI
One of Huang’s strongest arguments for continued growth is Nvidia’s position across almost every layer of the AI industry.
The company supplies infrastructure to major AI developers, including OpenAI, Anthropic and Google, while also supporting companies building open-weight models. Beyond the model developers themselves, Nvidia works with cloud providers, OEMs, neocloud companies, startups and data center operators.
That gives Nvidia visibility into where new AI infrastructure is being built and how quickly demand is developing.
Huang said Nvidia tracks everything from memory suppliers to data center construction and available power capacity around the world. In his view, this network provides the company with an unusually broad picture of future AI infrastructure demand.
“Nvidia runs every model,” Huang said, emphasizing the company’s role as a foundational platform for the AI industry.
The Company Is Also Investing in Its Customers
Huang was also questioned about Nvidia’s growing investments in companies that subsequently purchase Nvidia hardware — a strategy that has raised concerns about so-called circular financing.
His response was straightforward: Nvidia invests only when it believes there is genuine underlying demand.
According to Huang, companies receiving Nvidia investment are expected to have real customer contracts and revenue opportunities. He said Nvidia has identified roughly $100 billion worth of such contracts across these businesses.
The strategy allows Nvidia to help accelerate the development of AI infrastructure while potentially creating additional demand for its own computing platforms.
Competition Remains the Biggest Question
Despite Huang’s confidence, Nvidia’s dominance is not guaranteed.
Amazon, Microsoft and Google are developing their own AI accelerators, while companies such as OpenAI and Anthropic are increasingly exploring custom silicon. Nvidia also faces specialized chipmakers and newer competitors targeting AI workloads.
The larger question is whether AI infrastructure spending can continue growing at its current pace.
As AI companies mature, they are likely to become more efficient in how they use compute, models and tokens. That could eventually slow the growth of raw infrastructure demand.
For now, however, Nvidia remains at the center of the AI buildout. Its exposure stretches from chips and networking to complete data center systems and the companies developing the next generation of AI applications.
Huang’s 70% growth forecast is therefore less a prediction about a single product and more a bet that the AI infrastructure boom still has significant room to run.
Leave a Reply