Nvidia Says Revenue Could Grow 70% Next Year. Here’s Why

Nvidia CEO Jensen Huang says the company could grow revenue by around 70% next year, potentially taking annual sales from roughly $400 billion this fiscal year to about $680 billion. His reason is simple: AI demand is still growing faster than Nvidia can supply it.

Nvidia Says Demand Is Still Accelerating

Speaking at Goldman Sachs’ technology conference, Huang said Nvidia is no longer simply selling individual chips. It is supplying entire AI computing systems.

One of those systems, combining 36 Grace CPUs and 72 Blackwell GPUs, costs around $8.5 million, contains roughly 2 million parts, and requires about 250,000 kilowatts of power. Orders for that product are currently growing about 27% month over month, according to Huang.

Nvidia’s latest results already show the scale of demand. Second-quarter revenue reached $96.2 billion, up 106% year over year, while Data Center revenue jumped 117% to $89 billion. The company expects around $108 billion in revenue this quarter.

Why Huang Is So Confident

Huang argues that Nvidia has unusually good visibility into future AI spending because it works across almost every part of the industry.

Its hardware powers models from OpenAI, Anthropic, Google and Meta, while Nvidia also works closely with cloud providers, AI startups, data-center operators and suppliers.

“We’re tracking every single gigawatt of land, power, shell around the world,” Huang said, arguing that those relationships allow Nvidia to see infrastructure demand before much of the market does.

Demand may actually support even faster growth. Nvidia said customer forecasts point to its business potentially doubling next year, but supply constraints mean the company itself expects closer to 70% growth.

Competition Is Still Growing

The outlook comes despite increasing competition. Amazon, Microsoft, Google, OpenAI and Anthropic are developing their own AI chips, while companies such as Cerebras and newer semiconductor startups are challenging Nvidia directly.

Supply is another constraint. Rising memory costs and component shortages are expected to push Nvidia’s gross margin down toward 71% to 72% later this fiscal year.

Why it matters: A 70% revenue increase from a company already generating hundreds of billions of dollars in annual sales would be extraordinary. Nvidia’s argument is that AI computing is moving from an experimental technology into infrastructure that businesses can actually monetize.

The bigger question is whether AI spending can keep expanding fast enough to support Nvidia’s enormous growth expectations as competition, supply constraints and infrastructure costs increase.