A tangled network of Stoks Ai Balloon is afraid in the silicone valley

In Devday of Openai This week, Openai chief himself Altman himself was doing what American technology bosses rarely do these days: he actually answered questions from reporters.

“I know it’s tempting to write the story of the bubbles,” G -n Altman told me as he was sitting, surrounded by his best lieutenants. “There are actually many parts of AI that, in my opinion, are something like a bubble right now.”

In the Silicon Valley, the debate about whether AI companies are overestimated, has taken a new emergency.

The skeptics are private – and some now publicly – they ask if the rapid increase in the value of AI technology companies can be at least in part, the result of what they call “financial engineering”.

In other words, there are concerns that these companies are overestimated.

Mr Altman said he expects investors to make some bad calls and stupid startup companies will go with crazy money.

But with Openai, he told me, “Something real happens here.”

Not everyone is convinced.

In recent days, AI Bubble warnings come from the Bank of England, the International Monetary Fund, and JP Morgan boss Jamie Dimon, who told BBC “The level of insecurity must be higher in the minds of most people.”

Here, too, in what is often considered the technological capital of the world, fears are increasing.

At a panel discussion in the Silicon Valley computer history this week, the early II entrepreneur Jerry Kaplan told a packed audience that he had survived four bubbles.

He is particularly concerned now, given the scale of the money on the table compared to the Dot-Com boom. There is a lot more to lose.

“When (the balloon) breaks, it will be really bad, not just for the people in AI,” he said.

“This will entice the rest of the economy.”

However, at the Stanford Business School, which has been reported by its fair share of technology entrepreneurs, Prof. Antata Adthati says that while there are many attempts to model when we are in the balloon, it can be a useless exercise.

“It is very difficult to extract a balloon,” Prof. Addatti told me. “And you can’t say for sure that you were in one while the balloon burst.”

But the data is related to many.

Ai-connected businesses have represented 80% of the stunning profits on the US stock market this year – and Gartner estimates Global costs for AI Likely to reach a huge $ 1.5tn (£ 1,1TN) before it comes out 2025

Openai, who scored AI in the user mainstream with Chatgpt in 2022, is at the center of the tangled network of transactions that attract control.

For example, last month she made a $ 100 billion deal with a NVIDIA chipmaker, which in itself is the most prudent publicly traded company in the world.

It is expanding an existing investment that NVIDIA has already had in the company of G -N Altman – with expectations that Openai will build data centers powered by modern NVIDIA chips.

On Monday, Openai announced plans to buy equipment worth billions of dollars to develop AI by a competitor from NVIDIA AMD, in a deal that can make it one of the biggest shareholders of AMD.

Remember that this is a private company, even though such Recently rated half trillion dollarsS

Then there is a technological giant Microsoft, which is heavily invested, and Cloud Computing Behemoth Oracle also has a $ 300 billion deal with Openai.

Openai’s Stargate Project In Abilin, Texas, funded with the help of Oracle and the Japanese conglomerate Softbank and announced in the White House during the first week of President Donald Trump, is increasing every few months.

And as for NVIDIA, it has a share in the AI ​​Startup Coreweave – which supplies Openai with some of its massive infrastructure needs.

And as these increasingly complicated financing arrangements are becoming more and more frequent, experts here in the Silicon Valley say they can blur perceptions of demand for AI.

Some people also do not kill their words about it, calling the “circular funding” transactions or even “suppliers financing” – where a company invests or borrows to its own customers so that they can continue to make purchases.

“Yes, investment loans are unprecedented,” I told me Monday.

But, he added, “it is also unprecedented for companies to grow so quickly revenue.”

Openai’s revenue is growing rapidly, but they have never made a profit.

And it is hardly a good sign that the people I have said that they continue to lift Nortel – the Canadian telecommunications manufacturer that has borrowed a lot to help financing deals for their customers (and thus artificially enhance the demand for their goods).

For its part, Jensen Huang from NVIDIA defended its deal with OPENAI on CNBC on Monday, stating that the company is not required to buy its company’s technology with the money it invests.

“They can use it to do anything they like,” Huang said.

“There is no exclusion. Our main goal is to simply support them and help them grow – and grow the ecosystem.”

D -n Kaplan says he sees Several Telltale signs the AI ​​sector – and therefore the wider economy – may be in difficulty.

At Frothy Times, according to him, companies announce major initiatives and plans for products for which they still do not have capital.

Meanwhile, retail investors are trying to get into startup action.

AMD shares this week may indicate that investors are trying to get part of the Chatgpt wealth machine – and while all this is played out, a real physical infrastructure is being built aimed at satisfying the seemingly insatiable hunger for more AI development.

“We are creating a new ecological disaster created by man: huge remote data centers such as deserts that will rust and blow bad things in the environment, with no one left to be responsible because builders and investors will be disappeared long ago,” said Mr. Kaplan.

But even if we are in a balloon, the hope of the Silicon Valley is the investments that are being made now, it will not be necessary to waste.

“The thing that comforts me is that the Internet is built on the ashes of the over -investment in the telecommunication infrastructure of yesterday,” said Jeff Budier, who builds products in the AI ​​Community hug.

“If the infrastructure has an exceeded investment for AI loading, there may be financial risks tied to it,” he said.

“But it will allow you to many great new products and experiences, including those we don’t think today.”

There are many believers in AI’s potential to transform society.

The question is whether The money to finance the ambitions of the most important companies in the sector may dry out.

“Nvidia looks like the last lender or investor,” said Richard Jark, who founded the Uncoveralpha newsletter.

“Who else has the capacity of currently investing $ 100 billion in another company?”