How circular financing in the AI industry and the race to develop artificial intelligence are creating new risks for financial markets and individual investors.
By: Natalie Sarff
10 November, 2026
Artificial intelligence has quickly become one of the largest investment opportunities in current financial markets. Companies are committing hundreds of billions of dollars to developing AI models, building data centers, and manufacturing semiconductors needed to power them. As a result, AI has attracted significant investor attention, pushing stock prices and company valuations higher.
This excitement is not necessarily misplaced. AI has already demonstrated its ability to transform industries, improve productivity, and create entirely new business opportunities. However, there is a difference between believing a technology will be successful and assuming every company involved in its development will be profitable. This is where concerns about an AI bubble begin.
The risk is not just that AI companies might be overvalued, but that the financial relationships connecting them could create vulnerabilities throughout the market. This situation has many similarities to the dot-com bubble in the late 1990s that are worth paying attention to.
The Financial Structure Behind the AI Boom
One of the biggest concerns surrounding the AI boom is something called circular financing. This happens when a company invests in another business that then uses the money to purchase products or services from the original investor. For example, a semiconductor manufacturer might invest in an AI company, which then uses some of that funding to purchase the manufacturer’s chips. The manufacturer benefits both from its investment and the additional sales. This is not inherently a bad business decision. But it does make it harder to determine how much demand is coming from actual customers versus companies financially supporting one another.
NVIDIA and CoreWeave provide an example of this relationship. NVIDIA, one of the largest suppliers of AI semiconductors, invested $2 billion in CoreWeave in January 2026. CoreWeave provides cloud computing infrastructure for AI and relies heavily on NVIDIA technology. This means NVIDIA is both investing in and supplying a company whose growth creates additional demand for its products. Similar relationships exist between OpenAI, Microsoft, Oracle, and other technology companies.
The concern is what happens if one company struggles financially. If an AI developer cannot generate enough revenue to support its infrastructure spending, it may reduce chip purchases or delay data-center expansion. Its suppliers and business partners could then experience lower demand and potentially declining stock prices. This is an example of systemic risk, where financial problems at one company can spread to others through their business relationships.
Reports in April 2026 indicated that OpenAI had fallen short of certain internal revenue and user-growth targets. These reports raised questions about whether the company’s revenue growth would be sufficient to support its substantial future computing commitments. The IMF has also warned that interconnected AI investments could make market losses worse if investor expectations begin to weaken. For investors, understanding the difference between actual financial performance and future commitments is becoming increasingly important.
Similarities to the Dot-Com Bubble
The current AI boom has some major similarities to the dot-com bubble of the late 1990s. During that period, investors poured money into internet companies because they believed the technology would completely change how businesses operated. Many companies received extremely high valuations despite having little revenue or profitability. Eventually, expectations became too high. When companies struggled to generate profits, investor confidence declined, and the Nasdaq Composite experienced a dramatic collapse.
The internet itself was not the problem. It clearly went on to transform communication, commerce, and business. The problem was that investors overestimated how quickly companies would make money from the technology and how much those companies were actually worth.
There are some important differences between the two periods that are important to note. Many of today’s largest AI companies, including NVIDIA, Microsoft, and Amazon, already have established businesses, significant revenue, and existing customers. AI is also generating real commercial demand rather than relying entirely on the possibility of future adoption. Still, the dot-com bubble offers an important lesson for investors on overvaluing a company. Is this time reallydifferent?
Questions for Individual Investors
For college students and other individual investors, the AI boom raises questions about the relationship between growth potential and investment risk. Rapid company growth, major partnerships, and rising stock prices can create excitement around an industry, particularly when investors believe they may be witnessing the next major breakthrough. However, a company’s growth potential and its current valuation are two different things. Even a financially strong company with promising technology may have a stock price that already reflects high expectations for future earnings.
Another consideration is diversification. Owning several different AI stocks does not necessarily mean a portfolio is diversified if those companies depend on the same customers, suppliers, and sources of financing. If one major part of the industry struggles, several investments could be affected at the same time.
For younger investors with longer investment horizons, the AI boom presents an interesting question: How should investors balance the potential of a transformative technology with the uncertainty surrounding the companies developing it? Ultimately, how much of a company’s current value is supported by its financial performance, and how much depends on expectations for the future?