Deutsche Bank Warns AI Spending Surge May Be Propping Up Economy—But Bubble Risk Looms

Money continues to pour into the AI sector at an unprecedented pace, with Nvidia announcing a $100 billion investment in OpenAI and OpenAI itself planning massive new data center expansions even before its first $500 billion project is complete. Yet, as the frenzy grows, so does concern about sustainability. According to a research note from Deutsche Bank, the pace of spending may be reaching dangerous levels, raising red flags about a potential AI bubble. George Saravelos, a researcher at Deutsche Bank, warned that investment in AI is growing “parabolic,” so rapidly that it may now be the primary driver of U.S. economic growth. “AI machines—in quite a literal sense—appear to be saving the U.S. economy right now,” he wrote. Without the surge in tech-related capital spending, the U.S. economy would likely be close to, or already in, recession this year. This assessment aligns with data from the Wall Street Journal, which reported that AI-related capital expenditures have contributed more to U.S. economic growth this year than all consumer spending combined. At the heart of this trend is Nvidia, whose chip sales have become a linchpin for the entire tech cycle. Saravelos noted that the company is “currently carrying the weight of U.S. economic growth,” a burden that can only be sustained if its growth remains endlessly exponential—an assumption few believe is realistic. The risks of such extreme concentration are becoming harder to ignore. Torsten Sløk, chief economist at Apollo, recently stated that equity investors are “dramatically overexposed to AI,” with the S&P 500 showing an extreme level of concentration in just a handful of tech stocks. The danger, he warned, is that a downturn in this sector could trigger a broader market collapse. Compounding the concern is a report from Bain & Company, which found that AI’s demand for computing power is growing at more than twice the rate of Moore’s Law. By 2030, meeting that demand could cost $2 trillion annually. Yet, the world is currently $800 billion short of the necessary investment to keep up. Even more troubling is a study from MIT earlier this year, which revealed that only 5% of businesses adopting generative AI tools have achieved “rapid revenue acceleration.” The vast majority have seen little to no measurable benefit, raising serious questions about the return on investment across the industry. With such a massive financial commitment riding on uncertain outcomes, the AI boom may be less about proven innovation and more about momentum. As Saravelos put it, “In order for the tech cycle to continue contributing to GDP growth, capital investment needs to remain parabolic. This is highly unlikely.” The question now isn’t whether AI will transform the world—but whether the financial system can survive the reckoning when the bubble finally pops.