How OpenAI & the AI economy could collapse.

“The Mag 7 & OpenAI are artificially raising their valuations, causing an illusion effect for investors.”

The recent rally in the “Magnificent Seven” has raised concerns that some of their growth may not be as genuine as it appears. A big part of this comes down to a practice known as circular deals. This is not a new concept. It was a common feature of the dot-com bubble in the early 2000s, when companies boosted their revenues by cycling money through closed loops that created the illusion of demand.

A circular deal happens when money flows in a circle — going out one way and coming back another — making it look like real growth is taking place when, in fact, no new value is being created. 

Take NVIDIA as an example. On September 22, reports suggested the company was ready to invest around USD 100 billion into OpenAI. In return, OpenAI committed to purchase between USD 300 billion and USD 500 billion worth of NVIDIA chips. On the surface, this makes NVIDIA’s revenue growth look impressive. But the real question investors should ask is how much of that demand is authentic and how much is simply manufactured through financial engineering.

This dynamic is not unique to NVIDIA.

• Microsoft invested USD 13 billion into OpenAI, which then committed to spending USD 10 billion on Microsoft’s Azure cloud services. 
• Oracle has pledged funding for the USD 400 billion “Stargate” AI project in Texas, and OpenAI subsequently agreed to purchase USD 300 billion of compute power from Oracle.
• Similarly, Amazon invested USD 8 billion in Anthropic, which in return has chosen to run workloads on AWS.

What are the Best Tech ASX Stocks to invest in right now?

Check our buy/sell tips

At first glance, investors may see these OpenAI deals as a positive. Big revenue growth and rising demand. But in reality, what many of these companies are doing is using excess cash and liquidity on their balance sheets to recycle money back into artificial revenue growth, keeping “the music playing” for as long as possible. The result is an illusion of momentum that pushes stock prices higher, even though the demand isn’t entirely organic.

OpenAI has already committed more than USD 1 trillion to AI infrastructure spending, funded by major backers such as Oracle, SoftBank, and Microsoft. The question investors need to ask is: how will those commitments be paid back? This isn’t just about NVIDIA. The same patterns are showing up across the other “Magnificent Seven” names.

The danger of circular deals for shareholders 

• Illusory for shareholders
• Raises antitrust concerns

How does this relate to the 2000s .com bubble?

In the early 2000s, the Dow Jones Technology Index, essentially the “AI index” of that era, lost USD 2 trillion in market value over 2 years, a 75%+ percent collapse. The crash was the result of a combination of factors: overspending on technology investments, mismanagement of debt, and weakening demand as the hype cycle came to an end.

One of the more telling features of that late-stage bubble was the rise of circular deals. A prime example was AOL (America Online), the dominant internet service provider in the United States throughout the 1990s. AOL offered dial-up internet access, email, chat rooms, and a curated online portal at a time when the internet was still new for most households. By the end of the decade, AOL had tens of millions of paying subscribers and appeared to be a high-growth, high-margin internet powerhouse.

But when AOL’s growth projections began to slow, the company turned to a different strategy to keep the stock price elevated. It would invest in small e-commerce companies, take an equity stake, and, in return, those companies were expected to spend the invested money back on AOL’s services. On paper, this boosted AOL’s revenues, but in reality, it was a circular flow of capital designed to create the illusion of growth.

The problem, of course, was that this growth wasn’t backed by genuine demand. These deals inflated AOL’s numbers, but they didn’t create sustainable value. Eventually, when the dot-com bubble burst, the illusion collapsed. AOL’s stock price fell by more than 80 percent, and the company became a case study in how hype-driven strategies can unravel once the “music stops.”

What Investors Need To Know Now

“History doesn’t repeat itself, but it often rhymes.”

This quote captures our view on current tech valuations and the signs that point to a late-stage growth cycle in what could be a bubble. For investors, the key is caution. When you see price surges like Oracle’s recent move to USD 405, it is important to ask whether that growth is truly organic or being fueled by artificial numbers.

It seems that, unlike the Dotcom bubble, the AI potential bubble is much more concentrated in a few really big companies. In the Dotcom era, there were literally hundreds/thousands of companies investing in that space.

If a bubble were to pop, or if a recession were to hit, buying stocks at all-time highs on the back of inflated figures could prove damaging to your portfolio. That is why it is critical to take the time to do proper research and understand your investments, particularly in the fast-moving world of AI.