Nvidia CEO Huang says AI has to do ‘100 times more’ computation now than when ChatGPT was released

Nvidia CEO Jensen Huang said next-generation AI will need 100 times more compute than older models as a result of new reasoning approaches that think “about how best to answer” questions step by step.

“The amount of computation necessary to do that reasoning process is 100 times more than what we used to do,” Huang told CNBC’s Jon Fortt in an interview on Wednesday following the chipmaker’s fiscal fourth-quarter earnings report.

He cited models including DeepSeek’s R1, OpenAI’s GPT-4 and xAI’s Grok 3 as models that use a reasoning process.

Nvidia reported results that topped analysts’ estimates across the board, with revenue jumping 78% from a year earlier to $39.33 billion. Data center revenue, which includes Nvidia’s market-leading graphics processing units, or GPUs, for artificial intelligence workloads, soared 93% to $35.6 billion, now accounting for more than 90% of total revenue.

The company’s stock still hasn’t recovered after losing 17% of its value on Jan. 27, its worst drop since 2020. That plunge came due to concerns sparked by Chinese AI lab DeepSeek that companies could potentially get greater performance in AI on far lower infrastructure costs.

Huang pushed back on that idea in the interview on Wednesday, saying DeepSeek popularized reasoning models that will need more chips.

“DeepSeek was fantastic,” Huang said. “It was fantastic because it open sourced a reasoning model that’s absolutely world class.”

Nvidia has been restricted from doing business in China due to export controls that were increased at the end of the Biden administration.

Huang said that the company’s percentage of revenue in China has fallen by about half due to the export restrictions, adding that there are other competitive pressures in the country, including from Huawei.

Developers will likely search for ways around export controls through software, whether it be for a supercomputer, a personal computer, a phone or a game console, Huang said.

“Ultimately, software finds a way,” he said. “You ultimately make that software work on whatever system that you’re targeting, and you create great software.”

Huang said that Nvidia’s GB200, which is sold in the United States, can generate AI content 60 times faster than the versions of the company’s chips that it sells to China under export controls.

Nvidia counts on billions of dollars of infrastructure spend annually from the largest tech companies in the world for an outsized amount of its revenue. The company has been the biggest beneficiary of the AI boom, with revenue more than doubling in five straight quarters through mid-2024 before growth decelerated slightly.