Big money has been pouring into AI projects across the world for quite some time now. It can’t help growing even bigger. Cumulative spending globally on data centres alone could top $30tn by 2050, according to a projection by PwC, almost matching the value of outstanding US Treasuries.
It "dwarfs” what was spent in the late-19th-century railroad, or late-1990s dotcom booms, even after adjusting for inflation, PwC said. AI infrastructure spending in the US is projected to reach approximately 1.9% of GDP -- over $581bn -- in 2026, according to estimates from Goldman Sachs. Longer term, it’s projected to average 3.6% of GDP annually through 2032 in broader infrastructure forecasts.
Global AI spending accounts for roughly 0.9% of global GDP in 2026, projected to rise to 1.4% by 2028. According to Goldman Sachs Research, total worldwide AI investments — encompassing data centers, infrastructure, and enterprise/private deployments — are expected to exceed $1tn, representing a substantial share of worldwide economic output.
But lurking behind the dizzying projections and huge outlays by AI companies, alongside sky-high valuations, lie assumptions about vast broad-based productivity gains and future profits with little evidence so far - or historical precedent - to be sure they can deliver, according to economists.
JP Morgan wrote in August that broad-based productivity gains in the US, which leads the AI race, "remain elusive”, raising questions about the sustainability of AI valuations. A Bain & Company study said productivity gains from existing markets would not be enough to justify current outlays and "entirely new markets must emerge to close the funding gap”, suggesting those could range from using AI-guided robots to developing new materials for batteries and semiconductors.
Few doubt the potential of AI to transform everything from work in an office to research labs, just as past revolutions shrank journey times from days to hours or connected the world at the touch of a keyboard. What seems more immutable is the math behind securing a return on investment or the deadlines for repaying loans, leaving economists to work out the implications for the global economy beyond the ups and downs of investment cycles.
"Historical precedent suggests that technology-driven booms often end when infrastructure buildouts cease to deliver sufficient returns,” according to JP Morgan.
For the US alone – which, according to some estimates, accounts for about three quarters of the global AI spending total – investment will run as high as about $9tn from 2025 to 2032, equivalent to spending 3.2% of US GDP each year, according to Columbia Business School economist Stijn Van Nieuwerburgh.
He estimates the US AI sector would need to generate about $3.55tn in annual revenue by 2032 to earn a 10% return on investment. It earns a fraction of that now. In a wider sense, there are two schools of thought. The first dismisses serial scare stories about the power of AI as merely hype to build excitement around upcoming initial public offerings, according to Mike Dolan, a columnist for Reuters.
The other is that AI development is now too big to fail strategically, and governments have no option but to press ahead if economic, financial and even military security is threatened.
Going forward, AI may not be all about weapons of mass destruction in the "Science-vs-Nature Frankenstein” conflicts. With ample guardrails, it clearly has numerous positives, from accelerating the search for disease cures to even clean nuclear energy. To be sure, the Big Tech and governments will fiercely resist falling behind in the race.