Is the AI Conveyor Belt of Capital About to Stop?

The American economy is a bit higher than a big bet on AI. Morgan Stanley Investor Ruchi Sharma Recently mentioned The money paid in AI investment is now about 5% of the United States GDP growth in 2021, and AI companies are responsible for growing 5% of American shares. So how bad it is that the most recent big deal among the AI ​​giants, the contracts that have driven the stock prices dramatically, look like a snake eating its own tail?

In recent months, Nvidia declared that it would be Invest $ 100 Billion in OpenAIOpenai declared that it would be Provide $ 300 billion to Oracle For computing power, and Oracle declared to buy it Chips of $ 40 billion From Nvidia. These companies do not take any stream charts to get the feeling that is just moving towards money in each other. But surely that is not happening … isn’t it?

It is a bit harder to get more assurance than you think.

Is it all round tripping?

Many of these agreements are mutually beneficial, on their faces. If everything is at a level, these deals can be round then they should be taken forward. RBC Capital Market’s analyst Ish Sixth Jaluria Gizmodo told Gizmodo that they could be a “low-powered world”, which allows the rapid development of models that can produce higher return in investment.

“The better the model we have, the more we can realize in the use of this AI that is not only strong enough to manage it,” he said. “If that happens and it can generate real [return on investment] For customers … it saves the actual expenses, potentially new earning production opportunities, and it creates net facilities from the GDP perspective. “

So as long as we keep the AI ​​breakthroughs and these companies decide how to cash their products, everything should be fine. What does not happen in the off opportunity, though?

Jaluria said, “If this doesn’t happen, if the true enterprise does not accept AI, it is all round tripping,” said Jaluria.

Round-tripping, generally speaking, is immoral to trade or transactions to a particular resource or organization artificially and usually implies illegal practice, it looks like it is actually more valuable and demand. In this case, it will be technology companies that are trying to appear that they seem more valuable than to remove the share price of big deal with each other.

So what can suggest that this money is performing anything other than serving like a hot air in a quick inflamed bubble? Jaluria said he was keeping an eye on the rapid development of models, progress in performance and taking the overall AI. “If it takes and uses the Enterprise AI to change any step function in such a way, it creates an advantage,” he said.

Whether or not it is currently happening is a kind of viewer’s eyes. Open has certainly shown progress in its technology. Its Sora 2 Video is in the publication of the generation model Reveal a new hell on the worldUsed to produce significant amounts Copyright breach And Incorrect informationThe However the latest version of the company’s flagship model, GPT -5, Underhalemed And failed to survive as expected after being published in August.

The rate of technology adoption is also somewhat of the Rorshach test. The company is proud 10% of the world Chatzipt is using, and About 80% of the business world Says that it is looking for how to use the technology. However, the initial accepters are not able to find too much utility. Ay Survey from the Massachusetts Institute of Technology95% of companies that have tried to integrate the generator AI equipment in their activities have created zero return for investment.

Where these investments are creating returns in the stock market. Which is obviously, these companies do not just resort to anxiety to increase the bottom of each other.

For example, take Oracle. Last month, the cloud supplier was a Quadrant By all traditional dentic indexes. It missed both its earning and earning earnings and its net income was flat for years. And still, the The share price has risenThe Reason: List of obligations of the remaining performance of the company – Financial Agreements that will provide revenue that has not yet been completed. There, the company showed a lot of growth, which has increased by 359% from the previous year, resulting in an estimated $ 455 billion.

That money is not yet real. Not The company has promisedIt claims that its Oracle Cloud infrastructure will rise from $ 20 billion to about $ 150 billion before the start of the 2030s. However for investors was enough to run Oracle’s share price enough Top spot on the list of the richest people in the worldIn short, Elon musk jumping.

OpenAI is either the nexus point or the vacuum of the center

Most of these committed incomes will come from the opeina, which has promised to buy $ 300 billion computing power from the company for more than five years. The clock of this Agreement does not start until 2027, but assume that it actually happens, it will be a The largest cloud computing deals In history

It is also most impossible based on where the currently involved companies are standing. It has promised Openai to supply the calculation, Oracle desire Report Electric power has to produce 4.5 GHz, more valuable energy for two hover dams. On the other hand, the OpenA will have to pay about $ 60 billion per year to fit the bill for the OpenA contract. This is currently Generates earning about 10 billion dollarsWhich is statistically, less than $ 60 billion.

You can see the same notice size with the recent OPNA agreement with the NVDia rival AMD. The correct details of the agreement have not been reported, but the Chipmaker AMD Expect to generate a few billion dollars For the next half decades, it sells its AI chips to the Openai. As part of the agreement, OPENY AMD got a swath of shares, the company has the option of buying up to 10%. Luckily for OpenAI, it is no longer a good time to get your hand in some AMD shares before the announcement of a big AI-related agreement. Organization The share price has risen to about 35% Following the announcement.

OPENCE agreed with these two recent contracts in books More than 1 trillion dollars of computing deals So far this year. It’s a lot to spend for any organization, but it is especially a lot for a fixed-private organization that just reports 10 billion dollars in the estimated revenue Also in 2025. Even with its most recent funding round, the company is currently priced at about $ 500 billion.

Most of this agreement is an emergency. For example, Nvidia investment in OpenAI is not actually $ 100 billion, but 10 gigawatts are the initial 10 billion dollars for the power of one gigawatt data center with $ 100 billion in the end. However, the price and evaluation of the shares certainly behave in such a way that they are set on the stone. And the Open is working that way too. The agency claimed that it would be more than 10x in the next few years and projects that would hit it $ 129 billion in 2029The

Capital conveyor belt

This type of potential inflamed revenue is a kind of thing that forced some people to think about dot com about bubble in the early 2000s, where we saw Companies like Commerce One Get $ 21 billion evaluation despite having no earnings. However, Professor Peter Atwar and Consulting Agency of the Department of Economics, William and Mary, the president of the Financial Institute, AI bubble sees a separate image: the fall of the housing market.

“What we saw on the top of the mortgage market is the belts of all these forces, the money flowing from one party to the other. And what you started to see was the subject of multiple relationships so that any participating system on the system works to continue the system to continue the system,” he told Gizmodo. “” In different ways, we are seeing the same developer web flow across AI space “

It creates some obvious problems. According to the theory that deals round, the wheels have to be turned around to move the whole thing forward. If any of them stops, the whole thing stops, because they are all so connected that no failure is really detached.

Atwater said that the types of metric-constituent deals dominated by the headline in the AI ​​space, the types of metric-constituent deals are nothing more than what happened in the mortgage industry in the 2007, where some financial commitments are required to meet some conditions.

“Each one of the bubbles’ frenzy is an extra commitment. The purpose of the additional committee is to be at risk of what you believe in the future will be an intense rare product.” What we find repeatedly is one of the first obligations to cut the promise after the conditions have changed. “

At the moment, these promises have the stomach. Not guaranteed to be there in the future if all these committed returns of investment are not implemented. Atwater says that the contracts that have been created in the market, the Equity Markets are priced at “a remarkable multiple” and the suppliers capable of supplying promised products pay these transactions. There is no guarantee that all these factors will hold.

Maths are already quite intriguing. As Tech Commentter Ed Gitron IndicatedMicrosoft, Mita, Tesla, Amazon and Google have invested about $ 560 billion in AI infrastructure in the last two years. They bring up $ 35 billion in combination with AI-related revenue. Opena’s promises are even bigger, the returns that are reasonably smaller.

The development of the company and the expansion of its services will not depend on any small part of the huge data center projects, for which the same amount of energy must be managed as needed Combined New York City and San DiegoEnergy Not even available at presentThe And, again, there is no guarantee that the last product once all these energy costs and data centers are built, will actually produce earnings.

“In the end, if you have no customers for the product, AI will have no place because these companies cannot continue for anything. Hearing lots of calls in the past few weeks, these companies have an obvious open question about how to make this money,” said Atwater.

For this moment, everyone is looking green, and hope is eternal springs. Until it is, no one will ask where the income is coming from. “At this point, the AI ​​sector is working forever they are doing them as they have a long time under which they can get it out and make money,” said Atwater. “As long as confidence is high, this whole ecosystem can imagine. If confidence is expected, they are expected to provide real-mayad performance in a very short frame.”

Unfortunately, if this happens, it will not only be these companies that give rise to failure. “You have to see it as a greater ecosystem. To talk about AI today, it means we have to talk about the credit market, we have to talk about the credit market. Wall Street and AI are a single animal,” Atwater said that very few companies currently have a big idea about the whole of the American economy.

Lots of investors are entering the AI ​​space, fear of being lost in a market that looks like it can only rise. However, why some of them are looking at these evaluation and climbing prices, if all these meanings are transferred only around, they are a little curious about what can happen to the real value of the companies they are betting.

“” Why? “,” Atwater said, “The last question is asked in a bull market.”