Alphabet’s 100-Year Bond Tests the Limits of the AI Debt Boom
The century bond, Alphabet’s debut issuance in British pounds, drew nearly 10 times demand for the £1 billion ($1.37 billion) offering, according to Bloomberg.
The century bond, Alphabet’s debut issuance in British pounds, drew nearly 10 times demand for the £1 billion ($1.37 billion) offering, according to Bloomberg.
The century bond, Alphabet’s debut issuance in British pounds, drew nearly 10 times demand for the £1 billion ($1.37 billion) offering, according to Bloomberg. The coupon priced at 120 basis points above 10-year gilts, signaling strong appetite from pension funds and insurers seeking long-duration assets to match liabilities.
While the deal was widely viewed as well executed, its structure — a corporate borrower locking in capital for a full century — has reignited debate over how far the AI investment cycle has pushed credit markets.
Alphabet last week signaled it expects 2026 capital expenditures to reach up to $185 billion, primarily to expand AI compute capacity and data center infrastructure. The sterling bond forms part of a broader borrowing campaign spanning dollars, euros, Swiss francs and pounds.
Alphabet joins a short list of sterling century bond issuers, including the University of Oxford, the Wellcome Trust, EDF Energy and the government of Mexico — entities typically associated with long-lived, quasi-sovereign stability rather than rapidly evolving technology businesses.
For some investors, the issuance reflects a rational response to demand. Pension funds, in particular, require long-dated assets to hedge future payout obligations, and a highly rated issuer like Alphabet provides diversification away from sovereign-heavy portfolios.
But others see a different signal.
Bill Blain, CEO of Wind Shift Capital, described the surge in AI-related borrowing as “off-the-historical scale,” pointing to what he called a broader “AI hyperscaler debt-fest.”
Alphabet’s rivals — including Oracle, Amazon, and Microsoft — are also ramping up infrastructure investment. Strategists estimate major tech firms could issue as much as $3 trillion in debt over the next five years to finance AI capacity.
With credit spreads near historic lows and equity valuations for tech companies hovering near all-time highs, some market watchers warn the environment carries late-cycle characteristics.
“A 100-year bond… you can’t get much more frothy than that,” Blain said, arguing that such long-dated issuance can sometimes act as a psychological marker near market peaks.
Century bonds remain rare in corporate markets, and investors locking in yields just above 6% are effectively betting that Alphabet can sustain profitability and relevance through decades of technological disruption.
Unlike sovereign debt — where governments can print currency to manage liabilities — corporate borrowers face competitive pressures, shifting technologies and execution risk.
Over a 100-year horizon, even dominant tech platforms must continually reinvent themselves.
For Alphabet, the bond also broadens its funding base beyond dollar markets, tapping into U.K. pension and insurance demand without oversaturating U.S. credit channels. Most analysts expect the company to hedge the sterling exposure back into dollars.
The issuance ultimately underscores the scale of the AI infrastructure race. Hyperscalers are competing not just on models and products, but on physical capacity — land, power, chips and data centers — all of which require unprecedented capital commitments.
Whether Alphabet’s century bond proves to be a savvy locking-in of cheap long-term funding — or a high-water mark in AI-driven credit exuberance — will depend less on today’s order book and more on whether the returns on AI infrastructure justify the borrowing decades into the future.
Jay Goldberg is a staff writer at TechEchelon covering technology, markets, and policy. He files the breaking news and deal coverage that move the publication's core desks.
More from Jay →SK Hynix fell more than 10% and Samsung dropped 7.1% as Asian chip stocks extended a Wall Street sell-off driven by fears of further Federal Reserve rate hikes and mounting concerns over the cost of financing AI infrastructure projects.
BlackRock CEO Larry Fink warned Wednesday that any moratorium on data center construction in the United States would trigger a recession, citing the central role AI infrastructure investment now plays in the broader economy.
Semiconductor stocks fell for a sixth straight session as AI bubble fears and rising oil prices dragged on markets, even as analysts and commentators point to strong underlying chip demand heading into a critical stretch of earnings and inflation data.
Written for readers who already know the basics — markets, AI, and the policy decisions that shape both.