A Top Wall Street Strategist Says AI Spending Could Destroy Capital. Alphabet's Ruth Porat Says He's Wrong

A public disagreement has opened up between one of Wall Street's most prominent equity strategists and one of Silicon Valley's top financial executives over whether the artificial intelligence spending boom will pay off. Chris Wood, Jefferies' global head of equity strategy, has argued that the current scale of AI investment risks ending in significant capital destruction in the United States, pointing to Chinese open-source models that are commoditizing large language models and could erode the returns tech companies are counting on. Ruth Porat, Alphabet's president and chief investment officer and the company's former chief financial officer, pushed back in a recent interview with Fortune editor-in-chief Alyson Shontell, arguing that demand for AI capabilities remains strong enough that the capital-destruction case does not hold up, and that companies building genuinely differentiated products will do well regardless of how commoditized the underlying models become.

The numbers behind the disagreement are large by any measure. Amazon, Alphabet, Microsoft, and Meta are together projected to spend more than $720 billion on capital expenditures this year. Alphabet alone has guided to $195 billion to $205 billion in capital spending for 2026, more than double the $91.4 billion it spent in 2025, and the company has said it plans to spend significantly more again in 2027.

That spending is increasingly being financed with debt rather than cash on hand. Alphabet's long-term debt has climbed to roughly $98 billion, more than four times what it was a year earlier, and the company posted negative free cash flow in the second quarter for the first time since it went public in 2004. Across the industry, outstanding private loans to AI companies now exceed $200 billion, a sign that the risk Wood is describing would not be contained to the tech giants' own balance sheets if it materialized.

Wood's concern about Chinese competition is not hypothetical. Moonshot AI, a Chinese AI lab, has said roughly 70% of its usage now comes from outside China, evidence that cheap, open-source Chinese models are already winning customers globally rather than staying confined to a domestic market. Separately, growth at Anthropic, one of the best-funded Western AI labs, appears to have flattened recently, a data point skeptics point to as evidence that demand may not be limitless.

Porat has leaned on history to make her case, drawing comparisons to both the dot-com bubble and the 2008 financial crisis to argue that overbuilding infrastructure is not the same thing as destroying capital, so long as real demand eventually catches up to the spending. Her bet, in short, is that differentiation, not the price of the underlying model, will decide which AI companies end up justifying the hundreds of billions already committed.