Global technology companies have ramped up debt issuance this year to record levels, as an intensifying race to build artificial intelligence capacity forces even cash-rich firms to borrow heavily to fund that investment.
According to Dealogic data, global tech companies issued $428.3 billion of bonds in 2025 through the first week of December. U.S. firms accounted for $341.8 billion, while European and Asian tech companies issued $49.1 billion and $33 billion, respectively.
Traditionally reliant on internal cash flows, large tech firms have increasingly turned to debt, as borrowing costs are low and investor demand is strong.
Michelle Connell, president at Portia Capital Management, said debt-funded AI capex reflects a structural shift, as rapid technological obsolescence and short chip lifespans force companies to reinvest continuously.
The heavy issuance, however, has begun to lift leverage and weaken coverage ratios for some firms, raising questions about how balance sheets would hold up if AI investments fail to deliver expected returns.Â
That said, the biggest tech firms are generally profitable, have large cash buffers and a number of them rank among the world’s most valuable by market capitalisation.
A Reuters analysis of more than 1,000 tech firms with market capitalisations of at least $1 billion shows their median debt-to-EBITDA ratio rose to 0.4 at the end of September, nearly double the level seen during the 2020 debt surge. While leverage remains below levels typically viewed as alarming, the increase suggests debt is rising faster than earnings, which can pose a risk if cash flows fail to keep pace.
The median operating cash flow-to-total-debt ratio also fell to a five-year low of 12.3% in the second quarter before recovering modestly later in the year.
Credit markets have begun to reflect rising investor caution. Five-year CDS spreads on Oracle have nearly doubled to 142.48 basis points over the past two months, while Microsoft’s spreads have climbed to about 35 basis points from around 20.5 at the end of September.
“I view this phenomenon as the result of an overheated marketplace that has created its own self-serving narrative — go big or go home in terms of stock price,” said Scott Bickley, an advisory fellow at Info-Tech Research Group.Â
“This is neither sustainable nor repeatable as a permanent shift in operating modes for the hyperscalers.”
(Reporting By Patturaja Murugaboopathy; with additional reporting by Gaurav Dogra in Bengaluru; Editing by Amanda Cooper and Alexandra Hudson)
