A look at the OPENAI network entangled from an officer

The CEO of Openai, Sam Altman, is everywhere.

His artificial intelligence startup, now estimated at $ 500 billion, has been value implementation in the tens to hundreds of billions of dollars with infrastructure partners, even if he continues to burn cash mounds.

These expenses stimulate the market.

NASDAQ and S&P 500 have increased to record the summits this week after Nvidia agreed to invest up to $ 100 billion in Openai. This follows an agreement of $ 300 billion between Openai and Oracle In July as part of The Stargate program, an infrastructure project of $ 500 billion which is also funded by Flexible bank.

His commitments do not stop there. Core On Thursday, it was agreed to provide Openai up to $ 22.4 billion in AI infrastructure, an increase compared to the $ 11.9 billion that she initially announced in March. Earlier this month, a flea manufacturer Broadcom said he had obtained a new client of $ 10 billion and that analysts quickly pointed out of Openai.

While Optai says that scaling is the key to stimulating innovation and future IA breakthroughs, investors and analysts are starting to raise eyebrows on the sums of the bug, as well as Openai’s dependence on a network of increasingly interconnected infrastructure partners.

Openai took a $ 350 million stake in Coreweave before its IPO in March, for example. NVIDIA offered its financial participation in OPENAI by participating in a financing round of $ 6.6 billion in October. Oracle spends about 40 billion dollars on Nvidia fleas to feed one of the Openai Stargate data centers, according to a May Financial Times report. Earlier this month, Coreweave revealed an order worth at least $ 6.3 billion from Nvidia.

And thanks to its $ 100 billion investment in Openai, NVIDIA will obtain equity in the startup and will generate income at the same time.

OPENAI should only generate $ 13 billion in revenues this year, according to the financial director of Sarah Friar. She told CNBC that technology booms need daring Paris on infrastructure.

“When the Internet started, people felt like:” Oh, we over-construction, there are too many, “said Friar. “Look where we are today, right?”

Altman told CNBC in August that he was ready to manage the company at a loss in order to prioritize growth and its investments.

But some analysts increase the red flags, arguing that the Openai agreement with NVIDIA recalls the funding models of suppliers who helped to burst the DOT-COM bubble in the early 2000s.

NVIDIA has been the largest winner of the BOOM of AI so far because it produces the graphic treatment units (GPU) which are necessary to form models and execute large IA workloads. Nvidia’s investment in Openai, which will be paid in several years over several years, will help the startup create data centers based on GPU.

“You don’t need to be skeptical about the promise of AI technology in general to see this ad as a disturbing signal on how space has become self-referential,” customers wrote Tuesday. “If NVDA must provide the capital that becomes its income in order to maintain growth, the whole ecosystem can be unbearable.”

Peter Boockvar, director of investments, at some point, BFG Wealth Partners, said that business names in the late 1990s sounded in his ears after the announcement of the Openai-Nvidia agreement.

A key difference, however, is that this transaction is “so much greater in terms of dollars,” he wrote in a note.

“For all this massive experience to work without causing significant losses, OpenAi and its peers must now generate huge incomes and benefits to pay for all the obligations for which they register and at the same time provide a return to its investors,” said Boockvar.

An Openai spokesperson referred CNBC to the comments of Altman and Friar this week, adding that the company was pursuing “a unique opportunity which requires an equal ambition at the time”.

The total amount of the calculation request could reach 200 narcotic gigawatts by 2030, according to the Bathroom & Company 2025 technological ratio. Building enough data centers to meet this planned demand would cost around $ 500 billion per year, which means that AI companies should generate an annual turnover of 2 dollars to cover these costs.

Even if companies throw all their weight behind the cloud investment and data centers, “the amount would still drop $ 800 billion less than the necessary income to finance full investment,” said Bain.

There is a clear battle to come, but Altman d’Openai postponed concerns on Tuesday, rejecting the idea that the wave of infrastructure spending is exaggerated.

“This is what it takes to deliver AI,” Altman told CNBC. “Unlike previous technological revolutions or previous versions of the Internet, there are so many infrastructures that are necessary, and it is a small sample.”

Yun Li and Mackenze Sigalos of CNBC at this report

WATCH: Sam Altman of Openai defends the expansion of Stargate while the request for IA Monte