AI Demand Is Booming, but Money Is Becoming the Problem

Two of the biggest CEOs in technology have spoken about AI.

One is Arvind Krishna of IBM. The other is Jensen Huang of Nvidia. And each described something that seemed completely different.

Krishna said companies are now scrutinizing what they actually get out of AI. Huang said demand is “off the charts.” But if we look a little more closely, both are reaching the same conclusion.

The problem is no longer demand. It is money.

IBM

IBM reported its quarterly results, and on paper, they were very good.

Revenue reached $17.1 billion, beating market expectations of $16.84 billion. Earnings per share came in at $2.95, compared with analysts' expectations of $2.87. Software revenue increased 8% to $7.8 billion. Red Hat also grew by 11%, and the company raised its free cash flow target.

And the stock? It fell approximately 7% following the news.

You might be wondering, “Why did it fall if the numbers were so good?”

There are two reasons.

The first is guidance. IBM maintained its forecast for full-year revenue growth at 4% to 5%. It did not raise that forecast, as investors had hoped.

The second is something Krishna himself said. He explained that customers are becoming “much more conscious of the actual return on all their AI spending.” He went a step further, speaking about “AI spending that has gotten completely out of hand, without real returns.”

Yes, you read that correctly. These are the words of the CEO of a company that sells AI.

What does this mean in practice? Put simply, companies have started asking a very basic question:

“Okay, we spent the money. What did we get in return?”

This is known as ROI, or return on investment. And when buyers start scrutinizing every dollar, the entire AI market becomes more anxious about the prospects of the companies selling these solutions.

The results were good. But the market focused on just that one statement.

The Other Side of the Equation

Of course, Krishna did not present this as bad news for IBM. Quite the opposite. He described it as a “tailwind” for his company.

You might ask, how does that make sense?

Because customers do not want to abandon AI. They want to bring it under control. They are looking for help redesigning their processes and putting the brakes on costs. And IBM wants to provide that help.

Krishna also offered companies some advice: choose a small number of AI applications and implement them properly at scale, rather than running dozens of small experiments simultaneously.

Huang and Insatiable Demand

And this is where Jensen Huang enters the picture.

The Nvidia CEO spoke at GTC Berlin, and his message was completely different.

He said demand for so-called neoclouds is “off the charts.” In other words, it is extraordinary.

What exactly are neoclouds? Put simply, they are companies such as CoreWeave and Nebius. They buy enormous quantities of chips and rent computing power to anyone who needs it for AI.

But there is a big catch.

Huang also made a very important point: “What is missing in the middle is the working capital to fuel growth.”

In other words, the customers are there. The money needed to serve them is not sufficient.

And the market took notice.

CoreWeave fell 10%, while Nebius dropped 7.23%, as investors worried about their enormous spending requirements and debt. Nvidia itself closed down 4.4%, just one day after gaining approximately 2.5%.

Let's also mention that Huang turned his attention to Europe. “We're trying to kick-start the AI economy in Europe,” he said. According to Bloomberg, he also criticized Europe for falling so far behind in the race.

What Are Both CEOs Telling Us?

This is where things become even more interesting.

On one side, we have the buyer. Krishna says companies are looking more carefully at the bill.

On the other, we have the seller. Huang says the queue of customers is enormous, but the financing is missing.

Contradictory messages? On the surface, yes. But fundamentally, they are saying the same thing.

Neither CEO said that nobody wants AI. Both talked about money. One discussed how much value companies are getting from their spending. The other discussed who will provide the capital needed to meet demand.

So the market's question has changed.

It is no longer, “Will AI change everything?” It is now, “When will AI deliver returns, and who will pay for it?”