NEW YORK — U.S. non-financial firms have sold over €60 billion in euro-denominated bonds in 2026, a new record that signals a structural shift in how technology companies finance AI infrastructure spending that has grown beyond what their balance sheets can absorb.
The surge is driven by capital expenditures for artificial intelligence that have outpaced internal cash flows. Hyperscalers—Alphabet, Amazon, Meta, Microsoft and Oracle—now rely on external debt markets to fund data center buildouts and chip purchases at a scale their operating cash cannot match.
Morgan Stanley estimates these five firms could issue an additional €50 billion in euros before year-end, underscoring the sustained financing pressure.
Amazon's €14.5 billion eight-part euro bond deal in March set the record for the largest euro corporate bond offering ever. On Sept. 9 alone, the highest number of U.S. bond issuers entered European markets in a single day. Amazon also debuted sterling bonds for the first time that day, while Uber made its initial euro appearance—a pattern reflecting broad currency diversification by U.S. corporates seeking to match spending across multiple regions.
The five hyperscalers collectively hold approximately €40 billion in outstanding euro-denominated bonds, nearly 10 percent of all new euro non-financial corporate issuance. This concentration is reshaping European credit markets.
Total AI-related capital expenditures are estimated between $650 billion and $725 billion for 2026, with projections exceeding $1 trillion by 2028. That trajectory explains why even the largest technology firms have turned to foreign bond markets.
The European Central Bank has flagged crowding-out risks, warning that heavy U.S. corporate issuance could compress yields across the European corporate bond market and raise borrowing costs for domestic firms seeking capital.
U.S. companies pursue "reverse Yankees"—bonds issued in foreign currencies—to lower borrowing costs, diversify funding sources, access new investor bases and match foreign currency expenditures. For tech firms with AI spending concentrated globally, the strategic logic is straightforward: finance where the money is spent.
