Ask most equity investors what could derail the AI trade, and you’ll hear about valuation multiples, chip demand cycles, or whether enterprise adoption justifies the spend. Ask a credit desk the same question, and you get a different answer entirely: the AI boom has quietly become one of the largest coordinated corporate financing events in modern history, and almost none of the risk sits where equity investors are looking for it.
The numbers are worth sitting with. Combined capital expenditure among the largest hyperscalers is on track to exceed $600 billion in 2026, a 36% jump from 2025, with roughly $450 billion of that earmarked specifically for AI infrastructure. Amazon, Microsoft, Google, and Meta are each individually spending more than $100 billion this year. Capital intensity — capex as a share of revenue — has reached levels in the mid-to-high 40s and even into the 50s at some of these companies, territory that would have been unthinkable for asset-light software businesses even three years ago. The critical inflection isn’t the size of the number. It’s that operating cash flow no longer covers it. For the first time in this cohort’s history, the hyperscaler capex cycle has outrun free cash flow generation, and the gap is being plugged with borrowed money.
Unlock the article to continue reading.
Trusted by 100,000+ investors. We won't spam you. See our Privacy Policy.
Email Verification Required
Thank you for subscribing! Please check your email inbox and confirm your subscription to access the full article content.
If you don't see the email, please check your spam folder.
