“We’re looking at something completely different.”

in the previous few weeks Nvidia’s In the third quarter earnings report, investors debated whether the market was in an AI bubble, worrying about the huge amounts of money being spent on building data centers and whether they would see a long-term return on investment.

During an earnings call with analysts on Wednesday, NVIDIA CEO Jensen Huang began his comments by denying that premise.

“People talk a lot about the AI ​​bubble,” Huang said. “From our perspective, we see something completely different.”

In many ways, Mr. Hwang’s remarks are to be expected. He led the company at the center of an artificial intelligence boom that saw the company’s market capitalization reach $4.5 trillion as demand for Nvidia’s graphics processing equipment soared.

Huang’s breaking of bubble talk is important because Nvidia counts all the major cloud providers. Amazon, microsoft, googleand oracle — as a customer. OpenAI, Anthropic, xAI, and metais also a major purchaser of Nvidia GPUs.

Mr. Huang has deep market insight, and on the conference call he presented a three-pronged argument for why we are not in a bubble.

First, he said that he is focusing on GPUs because AI is needed in areas such as data processing, ad recommendations, search systems, and engineering. This means that the old computing infrastructure centered around central processors will migrate to new systems running on Nvidia’s chips.

Second, AI will not only be integrated into current applications, but will also enable entirely new applications, Huang said.

Finally, Huang said “agent AI,” or applications that can run without significant input from the user, will be able to reason and plan, and will require more computing power.

Citing Nvidia as an example, Huang said it is the only company that can address three use cases.

“When considering infrastructure investments, consider these three fundamental dynamics,” Huang said. “Each will contribute to infrastructure growth in the coming years.”

On its earnings call, Nvidia reported better-than-expected revenue and profit and issued better-than-expected guidance. Last month, Huang predicted that the company’s AI chip sales would reach $500 billion from 2025 to 2026.

The company announced Wednesday that its backlog does not even include some recent deals announced this week, such as the agreement with Anthropic and an expanded agreement with Saudi Arabia.

“The numbers will grow,” Chief Financial Officer Colette Kress said on a conference call, adding that the company is on track to meet expectations.

Before Wednesday’s earnings release, Nvidia stock had fallen about 8% this month. Other stocks related to AI have been hit even harder. coreweave It plunged 44% in November. oracle decreased by 14%, Palantir down 17%.

Some of Wall Street’s concerns are related to the debt certain companies have used to finance infrastructure projects.

“Our customers’ financing depends on them,” Huang said.

As for Nvidia, investors have expressed concerns in recent weeks about how much of the company’s revenue is going to a small number of hyperscalers.

Last month, Microsoft, Meta, Amazon, and Alphabet all raised their forecasts for capital spending to ramp up AI, saying they collectively expect to spend more than $380 billion this year.

Huang said that even without a new business model, Nvidia’s chips will boost revenue for hyperscalers as they power recommendation systems for short videos, books and ads.

People will soon start to see what’s going on beneath the surface of the AI ​​boom, rather than “a simple view of what’s going on with capital expenditures and investments,” Huang said.

WATCH: Nvidia announces third-quarter win