Nvidia Chief Forecasts 70 Percent Revenue Surge as Megacluster Shipments Accelerate

Nvidia expects to expand its top line by 70 percent next year, pushing annual revenue toward $680 billion despite mounting competition from in-house cloud silicon and rival chipmakers. Jensen Huang, founder and chief executive of Nvidia, reaffirmed the aggressive projection during an appearance at the Goldman Sachs Communacopia + Technology conference, according to a report from TechCrunch.

With Wall Street projecting the company will finish its current fiscal year near $400 billion, hitting that target would represent an unprecedented run of revenue expansion at mega-cap scale. Huang pushed back on assertions that Nvidia remains just a component vendor, emphasizing that its primary offering is now massive computing systems. An integrated setup that combines 36 Grace CPUs with 72 Blackwell GPUs over NVLink costs approximately $8.5 million, requires 250,000 kilowatts, and is currently experiencing 27 percent month-over-month sales growth.

Huang attributed his confidence to the breadth of Nvidia's hardware footprint across the artificial intelligence sector. Because labs such as OpenAI, Anthropic, and Google train and run models on its stack, Nvidia gathers real-time deployment data across global data center shells, power allocations, and cloud vendors. That pervasive reach, Huang argued, gives the chipmaker clear foresight into upcoming capital expenditure.

Huang also addressed skepticism surrounding circular investment deals, where Nvidia provides venture backing to AI startups that subsequently buy its accelerators. He rejected comparisons to historical telecom build-out bubbles, noting that Nvidia verifies customer demand before backing companies and has visibility into roughly $100 billion in binding contracts. "Well, it's not circular because we put a little bit of money in, and a lot of money comes back," Huang said, adding that the firm aims to minimize risk.

The massive revenue guidance lands at a pivotal moment for the semiconductor industry. While hyperscalers like Microsoft and Amazon develop proprietary chips, and challengers like Cerebras and Etched target specialized workloads, Nvidia's full-system networking architecture and entrenched software ecosystem continue to give it a dominant moat. For now, enterprise demand for compute clusters remains robust enough to absorb record hardware volume, even as industry observers watch for eventual efficiency gains in model architectures.