Nvidia says 'demand for AI infrastructure continues to expand at an unprecedented pace' but after speaking to experts about it, I'm not so sure

Nvidia has just announced its financial results for the previous quarter, and as expected, things are going very well for the GPU behemoth. The company is almost entirely AI-focused these days—I know, tiny violins for us PC gamers—and in its earnings call, Nvidia EVP and CFO Colette Kress said that there’s still an awful lot of demand for it:

“Demand for AI infrastructure continues to expand at an unprecedented pace. The build out of AI factories is accelerating. The value of NVIDIA AI infrastructure is rising.”

This increased demand, of course, allows for increased pricing: “The price of renting an H100 has risen 20% year to date, while A100 cloud pricing is up nearly 15%,” says the company. “Benefiting from the versatility of our platform and continuous performance enhancements enhanced by our software stack, customers are generating profitable revenue beyond the depreciable life of their GPUs.”

Latest Videos From

Nvidia has split itself, at least for financial reporting purposes, into two very broad segments: data center and edge computing. The former includes hyperscale and “ACIE” (AI clouds, industrial, and enterprise), and the latter includes “devices for agentic and physical AI”, with gaming PCs and consoles also falling in that bracket.

Nvidia’s previous quarter boasted an 85% year-on-year revenue increase up to $82 billion, and the company expects continued growth moving forwards. This will, of course, depend on continued AI demand, which Nvidia is very bullish about. The company’s CEO Jensen Huang, says regarding hyperscale CapEx (capital expenditure), “there are trillion[s of] dollars this year; I have every expectation it is gonna grow from here.”

Nvidia CEO Jensen Huang on stage at CES 2026.

(Image credit: Nvidia)

However, Nvidia would be expected to keep the fans flamed, not least because any hint at the company levelling out into stable profits rather than continued rapid growth could cause a dip in share price.

It’s therefore probably best to look elsewhere for opinions on whether AI infrastructure demand will continue to “expand at an unprecedented pace” and have Nvidia continue its incredible growth. To this end, I recently asked a couple of supply chain experts what they think about the future of AI growth, and received mixed opinions.

On the one hand, Jonathan Colehower, Managing Director, Global Operations & Supply Chain Practice at UST, expressed some doubts:

“I personally believe that we’re going to see a flattening on AI, and I think the market is going to get some relief before the end of 2026—we’re going to see some relief on global demand. I think that the past six months have been kind of a heyday on AI and datacentres, and we’re going to see that begin to slow.”

Jensen Huang discusses the Vera CPU

(Image credit: Nvidia)

I asked whether the AI bubble will pop. He replied:

“I think it’s going to be more of a soft landing… Particularly in the US, the way that the residents are responding to datacentres is really powerful, and they’re pushing back and saying, ‘No, I don’t want you to put a datacentre here.’ So, like any other economy, it’s just going to go and find a place where it can go, it can settle, and they’ll put their datacentre somewhere.”

Colehower continued: “But I do believe, personally, that we’re going to see a backing off on AI investments. I think that a lot of companies have made big promises on potential or future investments, and I think that those are going to be tempered over the next six months.”

On the other hand, Exiger Senior Vice President of Product Level Intelligence Derek Lemke is more optimistic:

“I don’t see it. I don’t see [demand] slowing down. I think I’m seeing broad AI adoption across all the Fortune 500 companies right now, which is driving the demand for compute, which is driving demand for GPUs, which is driving the demand for all of the upstream materials and requirements that are needed to power AI, even just energy and water.

Nvidia data center

(Image credit: Nvidia)

“And we’re starting to see strategic investments within the United States and regions that can start to locally support those high energy and high technical infrastructure needs that support that industry,” Lemke continues.

“So, I think that shift around regionalisation and localisation, we’re also seeing that on the AI front right now, especially when we’re at the placement of some of these AI centres. And, you know, both the requirements directly from an energy expenditure perspective that’s needed to power that, as well as the key materials and inputs as well.”

I suppose it’s asking for a little too much of a crystal ball to get a definitive answer to how the future of the industry will pan out—and given the nature and newness of the market, it’s not surprising that different experts might have different opinions on how the market might go.

At any rate, the fact that different experts seem to hold different opinions on the matter should give us reason not to take Nvidia’s predictions as certain.