Companies are spending enormous amounts of money just to stay competitive in the AI race, and debt levels are climbing as a result. In the latest development, Amazon has secured a loan of approximately $17.5 billion from a group of financial lenders, as reported by Bloomberg. The institutions backing this loan include Citigroup, JPMorgan Chase, Wells Fargo, HSBC, and BofA Securities. This arrangement is structured as a delayed draw term loan, which means Amazon can access the funds gradually according to its own schedule rather than taking the entire amount upfront, providing greater flexibility in how and when the money is used.

This loan comes just two days after news surfaced that Amazon would also raise $14 billion through a Canadian bond sale, bringing its total new financing to roughly $31.5 billion within about 48 hours. Exactly how Amazon plans to deploy all this capital remains unclear. Reuters notes that the new loan will be designated for "general corporate purposes."

Amazon is far from alone in this strategy. To fund new AI infrastructure—such as specialized chips and data centers—companies are making unprecedented capital expenditures. Increasingly, businesses are borrowing money to finance their massive AI buildouts. The question that investors and analysts are now asking is not whether this spending is necessary, but whether the returns will ever justify it.

The scale of this borrowing is striking, even by Silicon Valley standards. About a week ago, Google parent company Alphabet announced plans to raise $80 billion through a stock sale, intended to "fund its investments in a balanced way while retaining a healthy balance sheet." Meta has also revealed plans to raise $30 billion in a bond sale—its largest ever.