IS ARTIFICIAL INTELLIGENCE BOOM NEW FINANCIAL BUBBLE?

BANJA LUKA, DECEMBER 29 /SRNA/ - Artificial intelligence today occupies the position that internet startups held in the late 1990s, said expert in international and economic policy Nemanja Plotan.

Political leaders, technology executives, and investors speak of a "revolution," a "new industrial era," and a "fundamental transformation of the economy," Plotan pointed out in an op-ed for SRNA, which we are publishing in its entirety:

History, however, teaches us that moments of peak optimism are precisely when the most uncomfortable question must be asked: are we witnessing a genuine technological transformation - or yet another financial bubble?

Economist Jared Bernstein, former Chair of the White House Council of Economic Advisers, argues that there are serious reasons to view the current financial surge surrounding artificial intelligence through the lens of a bubble. Not necessarily a bubble that will burst tomorrow, but one whose core characteristics are already clearly visible.

A financial bubble emerges when investment in a particular asset class becomes so excessive that investors begin to doubt whether they will be able to achieve a reasonable economic return that justifies the capital invested. Once this confidence erodes, capital withdrawal leads to a sharp decline in valuations, bankruptcies, job losses, and, in some cases, recession itself. In the 21st century, we have already experienced two such episodes: the dot-com crash of 2001 and the global financial crisis of 2008.

To assess whether a third financial bubble is now forming before our eyes, it is worth examining the current valuations of the world’s seven largest technology companies by market capitalization. Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla today have a combined market value exceeding USD 20 trillion—an amount greater than the annual GDP of nearly every economy in the world, except the United States and China.

However, sheer size is not a reliable indicator of whether a financial bubble exists. A more relevant metric is the price-to-earnings ratio. For these companies - particularly within their artificial intelligence segments - this ratio is extraordinarily high and increasingly difficult to justify based on real revenues.

In other words, their stock prices are driven by expectations rather than by the cash they currently generate, and the gap between those two figures is enormous. Investors are not paying for what artificial intelligence delivers today, but for what they hope it might deliver at some undefined point in the future.

According to available estimates, Microsoft, Amazon, Google, and Meta are expected to invest between USD 70 and USD 100 billion annually in artificial intelligence in 2025 through capital expenditures and research and development.

On the other hand, revenues from artificial intelligence remain modest. Microsoft leads with approximately USD 15 billion in annual AI revenue, while Meta generates only around USD 300 million. OpenAI, one of the emblematic symbols of the AI revolution, plans investments approaching USD 1 trillion, while its revenues amount to just USD 13–15 billion.

An additional source of concern is the emergence of so-called circular investments. Nvidia, the producer of key AI chips, invests in OpenAI, while OpenAI uses a significant portion of that capital to purchase Nvidia hardware. On paper, balance sheets appear stronger; in reality, capital is not coming from outside the system but is circulating within the same ecosystem. Such arrangements can inflate valuations and create the impression of explosive growth in the short term, but they do not resolve the underlying problem—the absence of a sustainable and large-scale revenue base.

The central challenge these companies face is limited monetization. Around 80 percent of users engage with artificial intelligence only occasionally, without making a meaningful financial contribution. The key condition for avoiding a bubble is a transition from a consumer-oriented model to an enterprise-level one. At present, only about 10 percent of companies have seriously integrated artificial intelligence into their business processes. Without a significant increase in adoption, expected returns on capital remain highly uncertain.

Nevertheless, the artificial intelligence boom is not identical to the dot-com bubble, which emerged in the late 1990s amid speculative investment in internet startups with weak or nonexistent business models and culminated in a stock market collapse in 2000–2001. Unlike that period, today's technology giants possess real and profitable core businesses that do not depend exclusively on artificial intelligence for survival. Yet it is precisely here that a subtle distinction lies: if artificial intelligence fails to meet expectations, the bubble may burst without widespread bankruptcies—but not without losers.

This would also have serious geopolitical consequences, given that the United States and China are currently engaged in a race for dominance in artificial intelligence. Contemporary geopolitical competition unfolds across multiple dimensions, with AI occupying an increasingly central role. As a result, any state that gains an advantage in this race will be better positioned to project dominance across other spheres, including economic, political, and military power.

A potential bursting of the AI bubble would therefore not produce purely market-based consequences, but could reshape the very structure of this strategic competition. Unlike the United States - where AI development is heavily dependent on financial markets, high valuations, and the patience of private investors—China's approach is far more state-centered and less exposed to short-term market shocks.

In the event of a sharp correction or a loss of investor enthusiasm in the West, China could find itself in a relatively stronger position, as its AI sector is grounded in long-term strategic planning, state support, and integration into industrial and security structures.

A bursting of the bubble would disproportionately affect American technology giants, whose valuations and investment capacity are directly tied to market confidence. Weakening these firms would not only generate losses for investors but would also erode one of the central pillars of American economic and technological power.

If an economic correction were to occur, the largest nominal losses would be borne by the wealthiest investors, who hold roughly 90 percent of total market capitalization. Historically, however, the most severe social consequences are always carried by those with the least economic resilience: workers and the lower and middle classes, through rising unemployment and economic slowdown. In such a scenario, China would gain a significant comparative advantage, further undermining global confidence in the American financial system.