The surge in U.S. Treasury yields has triggered a debate over whether record borrowing by technology companies to fund artificial intelligence infrastructure is putting additional pressure on the bond market. The 10-year Treasury yield climbed to around 5.25% in late September 2026, reaching levels not seen since 2007 as investors weighed stronger U.S. economic data, higher oil prices and expectations for Federal Reserve policy. At the same time, major hyperscalers have issued hundreds of billions of dollars in debt to finance AI data centers, chips, power infrastructure and computing capacity.

Capital Economics argues that these developments should not be treated as the same story. While the rapid expansion of AI bond issuance could eventually influence long-term interest rates and Treasury term premiums, the research firm says the latest rise in yields has been driven more directly by economic strength, inflation concerns and changing Fed interest rate policy. Research from PIMCO, the Dallas Fed and BNY adds important context, showing that AI financing may affect bond markets through several channels without necessarily being the main cause of the current Treasury selloff.

Capital Economics argues that the recent rise in U.S. Treasury yields has more to do with higher oil prices, resilient economic growth and changing expectations for Federal Reserve policy than with the surge in AI bond issuance. The 10-year Treasury yield climbed to around 5.25% in late September 2026, reaching levels not seen since 2007 and prompting investors to look for an explanation for the sharp bond-market selloff. Capital Economics says record borrowing by AI-related companies has attracted attention, but the timing and scale of the Treasury move point more strongly to macroeconomic factors.

Economic Strength

One important driver has been unexpectedly strong U.S. economic data. S&P Global's flash U.S. Composite PMI rose to 58.4 in September, its highest reading since July 2021. Stronger activity reduces pressure on the Federal Reserve to lower interest rates and can strengthen expectations that policy will remain restrictive for longer. At the same time, elevated oil prices have added to inflation concerns, giving bond investors another reason to demand higher yields on longer-term government debt.

AI Debt

That does not mean the AI debt boom has no effect on bond markets. Capital Economics noted separately on September 24 that corporate issuance from AI-related companies has risen sharply, particularly at longer maturities. If borrowing continues at this pace, the additional supply could eventually put upward pressure on the Treasury term premium as corporate bonds compete with government debt for investor capital. Its assessment, however, is that this effect has probably been limited so far.

Another reason the direct Treasury impact may remain limited is that hyperscaler bonds do not compete with government debt on identical terms. Corporate bonds generally offer higher yields to compensate investors for credit and liquidity risk, while Treasuries play a different role as highly liquid benchmark securities and collateral across global financial markets. That means a surge in AI-related corporate issuance can reshape investor allocations without producing a one-for-one decline in Treasury demand.

PIMCO Evidence

Other research points in the same direction. PIMCO examined six unexpectedly large AI-related debt deals and found no statistically significant increase in 10-year Treasury yields around those issuance events. It also found little systematic effect on Treasury term premiums or swap spreads. PIMCO's conclusion is more nuanced than saying AI does not affect rates: the enormous AI capital spending boom may still push real interest rates higher by increasing demand for labor, power, construction and equipment. The evidence is simply much weaker for the narrower claim that AI companies are directly pushing Treasury yields higher by issuing bonds.

The AI bond boom has grown rapidly in 2026 as major technology companies raise debt to fund data centers, chips, power infrastructure and other AI investments. According to a Reuters analysis of LSEG data published July 29, 2026, Amazon, Alphabet, Meta and Oracle issued about $194 billion of bonds through July 7, already 79% more than the roughly $108 billion they issued during all of 2025. Goldman Sachs expected total 2026 bond issuance from those companies plus Microsoft to reach about $250 billion, before rising to roughly $400 billion in 2027.

The borrowing surge reflects the scale of the AI infrastructure spending cycle. Goldman estimated hyperscaler capital expenditure at about $750 billion in 2026, compared with projected operating cash flow of roughly $778 billion, with debt issuance financing around one-third of that spending. The flood of new supply is already changing the corporate bond market: Reuters found that 78 of 91 hyperscaler bonds issued in 2026 were trading at higher yields than at issuance by July 28, while investor demand for new deals had weakened. That helps explain why the AI debt boom matters for financial markets, even if the evidence remains mixed on whether it is directly responsible for the latest rise in U.S. Treasury yields.

The rise in the 10-year Treasury yield is being driven mainly by a combination of persistent inflation concerns, stronger-than-expected U.S. economic activity and expectations that the Federal Reserve may keep interest rates higher for longer. The benchmark yield reached about 5.29% on September 29, 2026, its highest level since 2007, before easing later in the session. Markets have become more sensitive to economic data because stronger growth makes additional Fed tightening easier to justify, while elevated energy costs continue to complicate the inflation outlook.

Oil prices have been another important factor. Geopolitical tensions in the Middle East pushed crude prices sharply higher during September, raising concerns that more expensive energy could feed back into consumer prices and delay progress toward the Fed's inflation target. At the same time, September's flash U.S. Composite PMI rose to 58.4 from 56.0, reinforcing the view that economic activity remains strong enough to withstand tighter monetary policy. These forces pushed investors to demand higher yields on longer-dated Treasuries, while broader attention remained on how CPI affects crypto and other risk assets as inflation expectations shifted.

Fed expectations remain especially important because Treasury yields can move quickly when investors reassess the future path of policy rates. On September 29, New York Fed President John Williams signaled that policymakers could afford to wait before raising rates again, helping the 10-year yield retreat from its intraday high and lowering market expectations for an October hike. That reaction illustrates why the latest Treasury move cannot be explained by AI bond issuance alone: oil, inflation data, economic growth and Fed communication are currently having a much more immediate influence on the bond market.

Research from PIMCO, the Dallas Fed and BNY shows that the relationship between AI debt and Treasury yields is more complicated than a simple crowding-out story. All three recognize that the AI infrastructure boom is adding substantial financing demand to fixed-income markets, but they differ on how much that borrowing is affecting U.S. government bonds.

PIMCO Findings

PIMCO tested six unexpectedly large AI-related bond deals over the past year and found no statistically significant increase in 10-year Treasury yields around those issuance events. It reached a similar conclusion when examining Treasury term premiums and swap spreads. PIMCO argues that AI spending can still contribute to higher real interest rates, but mainly because enormous investment in data centers, electricity, equipment and construction absorbs economic resources, not because corporate bonds are directly displacing Treasuries from investor portfolios.

Dallas Fed View

The Federal Reserve Bank of Dallas sees a clearer potential supply effect over time. Its February 2026 research estimated that AI-related investment-grade bond issuance could reach roughly $300 billion this year, creating as much as $360 billion of 10-year-equivalent duration, or about one-eighth of the duration supplied by U.S. Treasury issuance. The researchers also highlighted private-credit swaps and changes in financial-sector issuance as additional channels through which AI financing could add long-duration exposure to the market.

The Dallas Fed analysis also highlights why the maturity of AI financing matters, not just the headline amount raised. Long-dated corporate bonds add more interest-rate sensitivity, or duration, to the market than shorter-term borrowing. If AI companies increasingly fund infrastructure with 20- or 30-year debt, investors may need to absorb considerably more duration even when the dollar value of issuance changes only modestly. This is one reason the longer end of the bond market deserves closer attention as AI financing expands.

BNY Signals

BNY's market data suggests the pressure may already be visible at the margin. By late August, total U.S. investment-grade issuance had exceeded $1.5 trillion in 2026, with hyperscalers accounting for more than 12% of that supply. BNY also observed weaker long-end Treasury holdings and softer demand indicators as hyperscaler issuance accelerated, although it described the relationship as modest rather than conclusive. Taken together, the research suggests AI borrowing can influence long-term bond-market conditions, but the available evidence does not establish it as the dominant force behind the latest rise in Treasury yields.

The outlook for U.S. Treasury yields now depends less on the volume of individual AI bond deals and more on whether the broader AI investment cycle keeps the economy growing fast enough to sustain inflation and tighter monetary policy. After the 10-year Treasury yield reached 5.29% on September 29, 2026, markets remain highly sensitive to incoming inflation, employment and growth data. Capital Economics sees room for yields to eventually decline if monetary tightening proves less aggressive than currently expected, but the near-term path remains uncertain.

Fed Rate Expectations and the 10-Year Treasury Yield

Federal Reserve expectations are likely to remain the most immediate driver of Treasury yields. The Fed raised its target range to 3.75%–4.00% on September 16, while persistent inflation and solid economic activity have kept further tightening on the table. However, New York Fed President John Williams said on September 29 that policymakers could afford to wait before another increase, helping the 10-year yield retreat from its intraday high. Markets responded quickly, with expectations for an October rate hike falling substantially during the session. Upcoming inflation and labor-market data could therefore play a major role in whether yields remain above 5% or begin to ease.

How AI Spending Could Affect Long-Term Treasury Yields

The AI spending boom could still keep longer-term rates elevated through economic growth rather than bond issuance alone. U.S. core capital-goods orders excluding aircraft rose 1.6% in August and 10.6% from a year earlier, with strong business investment partly linked to AI infrastructure. Continued spending on data centers, chips, electricity and computing capacity could support economic demand and make inflation slower to cool. If AI capital expenditure weakens, that could reduce one source of growth and ease some of the pressure on long-term Treasury yields.

AI investment could also influence Treasury yields through productivity, although the direction is not straightforward. If spending on automation and computing eventually lifts productivity enough to increase the economy’s sustainable growth rate, real interest rates could remain higher even as inflation improves. But if AI projects produce weaker returns than companies expect, capital spending could slow, reducing investment demand and easing some of the pressure on long-term rates. The eventual effect will therefore depend on how much economic output the current AI spending cycle actually generates.

The debate over AI bonds and Treasury yields is ultimately about separating correlation from causation. Technology companies are borrowing at a historically rapid pace to finance AI infrastructure, and that additional supply can affect corporate credit markets and potentially influence long-term Treasury demand at the margin. But the evidence reviewed by Capital Economics, PIMCO and other institutions does not show AI bond issuance as the main explanation for the sharp rise in the 10-year Treasury yield.

For now, stronger U.S. growth, inflation risks, oil prices and expectations for Federal Reserve policy appear to have a more direct influence on Treasury yields. The longer-term picture is less settled. If AI investment continues to expand at hundreds of billions of dollars per year, its effect may increasingly appear through economic growth, real interest rates and duration supply rather than through individual bond deals. These shifts can also influence the Bitcoin price as investors reassess liquidity, interest rates and risk appetite across financial markets.

1. What is AI bond issuance?

AI bond issuance refers to debt sold by technology companies and other businesses to finance AI-related investment, including data centers, semiconductors, power infrastructure, networking equipment and cloud capacity. The term is informal because companies usually issue general corporate debt rather than bonds legally dedicated only to AI projects.

2. Why are major technology companies issuing more bonds for AI investment?

AI infrastructure requires unusually large upfront spending. Even companies with strong cash flow may use bonds to spread financing costs across many years, preserve cash for other investments and match long-lived assets such as data centers with longer-term funding.

3. Can corporate bond issuance compete with U.S. Treasuries?

Yes. Corporate bonds and U.S. Treasuries can compete for some of the same institutional investors, especially pension funds, insurers and asset managers seeking long-duration fixed-income assets. Higher corporate yields may attract capital away from government debt at the margin, although investor demand, credit risk and market conditions also influence those decisions.

4. What is the Treasury term premium?

The Treasury term premium is the extra return investors may demand for holding a long-term government bond instead of repeatedly investing in shorter-term securities. It can rise when investors face greater uncertainty about inflation, interest rates, government borrowing or the supply of long-duration bonds.

5. Why does the 10-year Treasury yield matter to financial markets?

The 10-year Treasury yield is a key benchmark for borrowing and asset valuations across the U.S. economy. Changes in the yield can influence mortgage rates, corporate financing costs, stock valuations and the discount rates investors use when pricing future earnings.

6. How can AI spending raise interest rates without companies issuing bonds?

AI investment can affect rates through the broader economy. Heavy spending on computing equipment, construction, electricity and skilled labor can increase demand for capital and resources, supporting economic growth and potentially keeping inflation or real interest rates higher even when that investment is financed with cash rather than debt.

7. Are AI bonds riskier than U.S. Treasury bonds?

They generally carry more credit risk because corporate repayment depends on the financial health of the issuing company, while U.S. Treasuries are backed by the U.S. government. Investors typically expect corporate bonds to offer a higher yield than comparable Treasuries to compensate for that additional risk.

8. Could slower AI spending push Treasury yields lower?

Potentially, but not automatically. A meaningful slowdown in AI capital expenditure could reduce business investment and economic demand, which might ease some upward pressure on long-term rates. Treasury yields would still depend on inflation, Federal Reserve policy, government borrowing and broader economic conditions.

This article is for informational and educational purposes only and does not constitute financial or investment advice. Market conditions can change rapidly, and readers should conduct their own research and consider their financial situation and risk tolerance before making investment decisions.