Jensen Huang reiterates Nvidia's 70% revenue growth forecast for next year

Nvidia repeats an unusually aggressive target

Nvidia founder and CEO Jensen Huang used an appearance at the Goldman Sachs Communacopia + Technology conference to repeat a forecast he first issued alongside last month's record quarterly results: that the company's revenue could expand around 70% year over year next year. According to TechCrunch, analysts expect Nvidia to finish its current fiscal year at roughly $400 billion, which would put the following year somewhere near $680 billion if the projection holds.

The remarks come at a moment when investors have been openly questioning how long Nvidia's run can last, given that its largest customers are also becoming its rivals.

Reframing the GPU as an $8.5 million machine

A central part of Huang's argument, as reported by TechCrunch, is that Nvidia should no longer be thought of as a chip vendor. He contrasted the $399 graphics cards of the company's gaming era with what a GPU means today: a machine he valued at $8.5 million, built from roughly two million parts, drawing 250,000 kilowatts and linked together with NVLink — hardware so large that shipping it requires cargo planes.

He also offered a concrete demand signal: a system combining 36 Grace CPUs with 72 Blackwell GPUs is currently seeing order growth of 27% month over month.

Visibility as the core thesis

Beyond current sales, Huang argued Nvidia has an information advantage that lets it forecast with confidence. Because its hardware trains and serves models from Anthropic, OpenAI and Google as well as open-weight alternatives, and because the company works with clouds, OEMs, neoclouds and AI-native startups, he claimed Nvidia effectively sees the whole buildout as it happens. He said the company is tracking every gigawatt of land, power and data center shell — the empty buildings before computing equipment is installed — around the world, and described Nvidia as a foundational platform for the entire AI industry.

Answering the circularity question

That breadth of involvement inevitably raised the subject of Nvidia's investments in companies that then buy its products — arrangements TechCrunch noted echo the vendor-financing practices that helped sink earlier infrastructure suppliers such as Lucent Technologies.

Huang was dismissive, joking that a dollar invested returns a hundred, and asking rhetorically whether anyone would object to that kind of circularity. More seriously, he said Nvidia only puts money in after confirming that a target has genuine customer contracts producing revenue — contracts he says he has seen to the tune of $100 billion — and framed the strategy as seeking certainty rather than taking risks.

Competition and caveats

The bullish outlook arrives amid mounting pressure on several fronts, TechCrunch points out: Amazon, Microsoft and Google are all designing their own AI chips, the labs Anthropic and OpenAI are pursuing their own silicon, and challengers such as newly public Cerebras and startup Etched are pushing alternatives.

There is also a fragility in the demand base itself. Huang acknowledged that much of the current spending surge comes from AI-native startups that raise large sums and immediately spend most of that capital on their own AI operations. As the industry matures and customers get more efficient with infrastructure and token usage, that spending pattern could shift.

Why it matters

Nvidia's guidance functions as the closest thing the AI economy has to a demand gauge. A 70% growth forecast to roughly $680 billion is a statement about data center construction, power availability, memory supply and the continued flow of capital into AI startups — not just about one company's sales.

The claim of total visibility is equally significant. If one vendor really does have line of sight into every major data center project and every model being trained, it holds a strategic position no competitor matches. But that same concentration cuts both ways: the forecast depends heavily on startups continuing to spend investor money on Nvidia gear, and history suggests dominant infrastructure players eventually face both disruption and tighter scrutiny of how their ecosystems are financed.