Over the past two years, artificial intelligence has transformed from a technological trend into one of the pillars of the global economy. Large technology companies are investing sums unprecedented in modern history, raising the question of whether AI is becoming a sector that is “too big to fail.” In other words, so fundamental that its collapse would threaten the entire financial system.
Last year alone, the largest technology companies spent approximately $400 billion on the construction of data centers, which are key to the operation of AI. This sum even exceeded consumer spending in the first half of 2025. And according to estimates, it is set to grow further in 2026.
This is not a marginal investment. Artificial intelligence has become the main driver of growth in technology stocks. Nvidia gained nearly 40 percent in 2025, Alphabet roughly 65 percent. “Even if the stock market declines, AI as a technology will continue,” says James van Geelen, founder and head of research company Citrini Research. According to him, this is not a bubble that will simply burst.
He believes that the real turning point will come when AI fully manifests itself in the labor market. “2026 will likely be the year when we start to see people losing their jobs and entire professions disappearing,” he warns. “From a societal perspective, this is more frightening than the fear that the technology won’t work.”
Until then, the main economic impact of artificial intelligence will be seen primarily in massive capital expenditures—that is, in infrastructure development. But this is where a new risk arises.
Debts that could shake the system
According to Patricia McCoy, a law professor at Boston College and an expert on government bailouts, concerns about “too big to fail” are primarily related to how AI development is financed.
“Some of these companies are borrowing to expand. They are taking out loans and issuing bonds,” she explains. Amazon, Meta, Alphabet, and Oracle all sold bonds last year to finance data centers. Meta, for example, issued $30 billion in bonds in October—the largest single issuance in the U.S. high-quality bond market for the entire year. The question is: will these companies be able to repay their debts?
“If any of them default on their debts, it could cause the collapse of at least one large financial institution that owes them money,” says McCoy. “And the failure of one bank or fund can trigger a domino effect.”
This domino effect is a typical feature of systemic risk – and a key part of the concept of “too big to fail.” So it’s not just about one technology company going under. It’s about the fact that its collapse could drag down a whole chain of financial institutions.
What does this mean for investors?
For ordinary investors, this creates a strange paradox. On the one hand, there is a sector that is driving markets up, attracting capital, and promising a fundamental transformation of the economy. On the other hand, its interconnectedness with the financial system is growing—and with it the risk that any problems will spill over into the entire economy.
Artificial intelligence today is not just a technological innovation. It is becoming an infrastructure in which banks, bond markets, and government institutions are entangled. And that is precisely why the question is being asked more and more often: is AI now too big to fail?
We do not yet know the answer. What is certain, however, is that the development of this sector is no longer just a story of technological progress. It is a story of debt, risk, and how deeply a single trend can penetrate the foundations of the entire economy.
Sources:
https://edition.cnn.com/2025/09/24/business/big-tech-nvidia-chatgpt-funding-nightcap
~:text=Chief%20among%20them%20was%20Meta,support%20its%20own%20AI%20capex.
https://www.marketplace.org/story/2026/01/22/is-ai-becoming-too-big-to-fail
