AI Hyperscalers Drive $2 Trillion Bond Boom, Pushing Up US Treasury Yields

3 articles · Updated · Fortune · Aug 28

$1.7 trillion in U.S. investment-grade corporate bonds had been issued through July, up 27% from a year earlier, with AI hyperscalers helping put 2026 on track to top $2 trillion.
That borrowing has not meaningfully widened corporate spreads because demand for AI-linked debt remains strong, so the market is adjusting through higher Treasury yields instead, according to Ed Yardeni.
$40 trillion in U.S. debt, a fiscal-year deficit nearing $2 trillion and annual interest costs around $1 trillion are already pressuring the Treasury market, with higher oil prices and a firm economy adding to yield pressure.
Foreign private buyers have purchased more U.S. corporate bonds than Treasuries over the past year, while officials including Scott Bessent and Kevin Warsh have flagged the scale of capital pouring into AI infrastructure.
S&P Global has warned of market fatigue as leverage rises at hyperscalers that were once seen as dependable cash generators, echoing broader concerns that AI spending is straining Big Tech balance sheets.