BlackRock CEO Larry Fink recently warned that a halt to data centre construction in the US could push the country into a recession, saying the economy increasingly depends on spending on artificial intelligence (AI) infrastructure. Now, Jefferies’ global head of equity strategy, Chris Wood, has raised concerns about the same investment boom, warning that cheaper Chinese AI models could threaten the returns American companies expect from their massive spending. The two views highlight a growing debate over the future of the AI boom. While Fink believes continued investment in data centres is essential to economic growth, Wood has warned that the money pouring into AI infrastructure could lead to massive capital destruction in the US if companies fail to earn enough from their investments.Wood’s concern comes as major technology companies continue to spend billions of dollars on AI chips, computing facilities and the power systems needed to run them. Amazon, Alphabet, Microsoft and Meta are expected to spend more than $720 billion on capital expenditure this year, according to analyst estimates compiled by Bloomberg.
Why Jefferies’ Chris Wood is worried about America’s AI spending
Jefferies’ Chris Wood has warned that the AI boom could destroy significant amounts of capital in the US over the long term. The scale of the spending helps explain the concern. As mentioned above, Amazon, Alphabet, Microsoft and Meta are expected to spend more than $720 billion on capital expenditure this year, with much of it going towards AI infrastructure.Alphabet alone expects to spend between $195 billion and $205 billion in 2026, more than double the $91.4 billion it spent in 2025. Alphabet President and Chief Investment Officer Ruth Porat has also indicated that the company plans to spend “significantly more” in 2027.But the investment is increasingly being funded through debt, including loans from private credit funds. Outstanding private loans to AI-related companies have risen to more than $200 billion by some estimates. Even Alphabet reported negative free cash flow in the second quarter for the first time since its 2004 initial public offering, while its long-term debt reached about $98 billion at the end of June, more than four times the level a year earlier.The bigger concern for Wood is whether companies will be able to earn enough from AI to justify this spending. Wood has argued that Chinese models could turn large language models into a commodity, making it harder for US companies to recover the enormous costs of developing and running them. He has also said China “has become a technological peer to the U.S. in AI.”
Why Larry Fink says stopping data centre construction could hurt the US economy
Larry Fink, on the other hand, has recently warned that a halt to data centre construction could have consequences well beyond the technology sector. Speaking to CNBC after BlackRock reported its third-quarter earnings, he said, “If we have a moratorium of some sort in this country, we will be in a recession.”“So much of our economy is based on the buildout of this capex,” he added, referring to capital expenditure on AI infrastructure.Data centres require spending on buildings, computing hardware and energy systems. This creates work for construction companies, equipment suppliers and businesses involved in power infrastructure. Fink argues that stopping new projects could therefore affect jobs and economic activity across several industries.Fink further argued that the economic benefits of building more AI capacity outweigh the concerns. Speaking at the Milken Institute Global Conference in May, he said, “I don’t believe we’re moving fast enough,” while dismissing concerns about an AI bubble.“There is not an AI bubble. There is the opposite. We have supply shortages. Demand is growing much faster than anyone has ever anticipated — and this is just a U.S. phenomenon,” he added.
What this may means for US economy
Fink and Wood are looking at different sides of the same investment boom. Fink believes that AI infrastructure spending supports jobs and economic growth, while Wood warns that the spending could destroy capital if the technology becomes cheaper and the returns fall short of expectations.Both outcomes could matter for the wider economy. A sharp slowdown in data centre construction could affect industries that depend on the investment. But if companies borrow heavily to build AI infrastructure and fail to earn enough from it, the losses could spread to investors and lenders.For now, major technology companies are showing little sign of pulling back. Alphabet’s Porat said, “You must invest for long term growth,” adding, “By the time you need it, it’s too late to say, OK, now we’re going to start.”The question is whether demand for AI will grow quickly enough to justify the enormous sums being spent, especially as Chinese models offer cheaper alternatives. That tension sits at the heart of the debate over whether America’s AI boom will sustain economic growth or leave companies facing costly investments that fail to pay off.