A new term, "circularity," is gaining traction among market analysts to describe a pattern of massive, interlocking investments in the artificial intelligence sector. Tech giants are pouring billions into one another, creating a self-reinforcing cycle of capital that raises questions about the true scale of market demand and echoes financial practices from the dot-com era.
Companies at the forefront of the AI boom, including Nvidia, OpenAI, Google, and Oracle, are engaged in complex financial arrangements where money flows from one firm to another, often to finance the purchase of the original investor's products. While legal, this trend is prompting a closer look at the foundation of the current AI market valuation.
Key Takeaways
- Major tech companies are investing billions in each other, a practice now termed "circularity."
- This pattern involves one company funding another, which then uses the capital to buy the first company's products.
- The practice is reminiscent of "round-tripping" deals that were common during the dot-com bubble of the late 1990s.
- Analysts are concerned this could be artificially inflating demand for AI infrastructure, creating a potential investment bubble.
The Multi-Billion Dollar Feedback Loop
The financial mechanics of the current AI boom are intricate. At its core, circularity describes a scenario where capital makes a round trip. For instance, a leading AI chip manufacturer might invest a significant sum in an AI model developer. That developer, in turn, could invest in data center providers who need to purchase massive quantities of chips from the original investor.
A recent example illustrates this dynamic. Nvidia, a dominant force in AI chip manufacturing, announced a major investment in OpenAI. Subsequently, OpenAI revealed large investments in companies like SoftBank and Oracle, both of which are building vast data centers. These data centers are primary customers for Nvidia's high-performance chips.
Effectively, a portion of the initial investment capital cycles back to the original company as revenue. This creates a powerful feedback loop where investment drives sales, which in turn justifies further investment. The concern for market watchers is whether this represents genuine, sustainable demand or an engineered boom funded by the industry's own key players.
Trillions in Projected Borrowing
The scale of investment is staggering. According to estimates from Morgan Stanley, a handful of major AI companies are projected to borrow approximately $1.2 trillion over the next three years to finance the build-out of AI data centers and infrastructure. This level of debt underscores the high stakes involved if market demand fails to meet expectations.
A Familiar Pattern with a New Name
For veteran investors, this scenario feels familiar. During the technology boom of the late 1990s and early 2000s, a similar practice was known as "round-tripping." It involved one company investing in another, which then used those funds to purchase services or products from the first, thereby inflating reported revenues.
Historical Context: The Dot-Com 'Round Trip'
In the dot-com era, round-tripping transactions were sometimes used to create a false impression of rapid growth. While not all such deals were improper, some crossed legal lines, leading to serious consequences for executives at companies like the former media conglomerate AOL Time Warner Inc. The practice became a hallmark of a speculative period where valuations often became detached from underlying profitability.
Today's circular investments are happening in a different context. The companies involved are often established, cash-rich giants. However, the fundamental question remains the same: is the industry building capacity far beyond what the market will ultimately support?
The risk is that if the anticipated explosion in consumer and enterprise AI applications does not materialize as quickly or as broadly as hoped, the immense infrastructure being built could become underutilized. This could lead to significant financial losses and a painful market correction, similar to what occurred when the dot-com bubble burst.
Is This Time Different?
Proponents of the current investment climate argue that direct comparisons to the dot-com era are flawed. Many of today's leading tech firms, such as Google and Meta, have robust, cash-generating core businesses that make them far more resilient than the speculative startups of the early 2000s.
Financial analysts at JP Morgan Chase have noted that while some caution is warranted, the underlying fundamentals of today's tech leaders are stronger. For these giants, even a significant overinvestment in AI might not pose an existential threat. Their vast resources allow them to make long-term bets on paradigm-shifting technology.
"While some caution is warranted, we think the better question is not whether today’s deals resemble the dot-com era, but whether the underlying fundamentals do," wrote JP Morgan Chase analysts Stephanie Aliaga and Nicholas Cangialosi.
However, the sheer scale of borrowing and investment introduces systemic risk. If the returns on these multi-trillion-dollar bets don't materialize as planned, the impact could extend beyond the tech sector and affect the broader economy.
The Unchanging Laws of Investment
There is little doubt that generative AI is a transformative technology. Its potential to reshape industries, create new markets, and change daily life is real, much like the internet was a generation ago. The internet's revolution produced some of the world's most valuable companies, but it also left behind a trail of failed enterprises that were either too early or simply couldn't find a sustainable business model.
The same will likely be true for AI. The technology itself is paradigm-changing, but that does not guarantee the success of every company participating in the current boom. The fundamental principles of investment remain unchanged: for an investment to be successful, it must eventually generate a financial return that justifies the initial outlay.
The current period of frenzied activity and circular financing makes it difficult for investors to distinguish between genuine growth and self-generated hype. Ultimately, the market will learn, as it always does, that even the most revolutionary technology is subject to the old-fashioned laws of financial gravity.