Mind the cap(ex): why sterling credit may be less exposed to the AI debt wave

Everyone knows artificial intelligence is transforming the technology sector. However, investors are now starting to appreciate the extent to which it is influencing corporate bond markets.

The world's largest technology companies, often referred to as hyperscalers, are in the midst of one of the most ambitious investment programmes in corporate history. Microsoft*, Amazon*, Alphabet*, Meta* and Oracle* are collectively committing hundreds of billions of dollars to data centres, semiconductors, networks and power infrastructure in pursuit of AI leadership[1]. While these companies continue to generate significant cashflows, the scale of their investment plans has increasingly led them towards debt markets for funding.

As a result, corporate bond markets are beginning to diverge between those absorbing the majority of AI-related issuance and those largely sitting on the sidelines. So far, sterling credit has fallen firmly into the latter category.

The AI arms race requires unprecedented amounts of capital

Much of the excitement surrounding AI has largely centred on technological innovation and potential productivity gains. However, from a credit market perspective, the more immediate story is capital expenditure.

Hyperscaler capital spending has accelerated sharply over the last two years as companies race to build the infrastructure required to support generative AI. While markets have traditionally viewed these issuers as infrequent borrowers with fortress balance sheets, they are increasingly becoming regular participants in corporate bond markets.

Issuance is no longer simply about taking advantage of low funding costs. Instead, debt markets are becoming an important strategic funding source for what appears to be a multi-year investment cycle.

As a result, supply is becoming an increasingly important driver of spread performance. In some parts of the market, investors are no longer focused solely on credit fundamentals, but are looking at future potential issuance. 

Not all bond markets are affected equally

A key feature of the AI financing boom is that the issuance requirement is not being shared equally across credit markets.

The US dollar market remains the natural home for hyperscaler funding. It offers unmatched depth, the largest execution capacity, the broadest investor base and the ability to issue across very long maturities. Unsurprisingly, this is where the overwhelming majority of AI-related debt financing has occurred to date.

Recent market data suggests that approximately $194 billion of global hyperscaler bond issuance has taken place year to date, of which roughly $182 billion has been issued in US dollars. Within US investment grade, hyperscaler net supply accounts for approximately 24% of overall net issuance.[2]

For investors, this matters. Persistent supply has created growing concerns around market absorption capacity, wider new issue concessions and the increasing dominance of hyperscalers within benchmark indices.

In short, US credit investors are increasingly financing the AI buildout.

Europe is participating, but selectively

The euro market has become a much more meaningful alternative funding source in recent years.

European investors are therefore participating in the AI funding story, but they are not carrying the same burden as their US counterparts. We covered this topic in a recent blog.  

Sterling remains the outlier

The sterling market is where the story becomes particularly interesting. Unlike the US, and increasingly the euro market, sterling credit has seen considerably less hyperscaler issuance. To date, the most notable example has been Alphabet’s £5.5 billion transaction.[3] We believe there are good reasons why.

First, market size matters. The sterling investment grade market is significantly smaller than either the US dollar or euro markets and is less suited to accommodating the enormous deal sizes increasingly associated with AI financing.

Second, funding economics continue to favour dollars. For issuers requiring tens of billions of dollars of annual financing, the deepest and most liquid market remains the most efficient source of funding.

Third, while issuers value funding diversification, there are limits. The scale of current AI investment programmes means that borrowers inevitably gravitate towards markets with the greatest capacity.

Given this, we believe the sterling market is unlikely to become the destination of choice for the largest AI-related financing transactions. Therefore, sterling's relative absence from the AI funding boom may prove more structural than temporary, in our view.

Could this be good news for sterling investors?

For investors, less supply is not necessarily a disadvantage. Credit spreads are influenced by three factors: fundamentals, valuations and technicals. While fundamentals often dominate market narratives, technical factors can become increasingly important when supply reaches extreme levels.

A market that avoids a significant proportion of global issuance potentially faces fewer forced buyers, fewer new issue concessions and less benchmark-related crowding.

There are widening spread differentials for hyperscaler issuers, growing investor fatigue and concerns about the unpredictability of future issuance. 

Sterling investors are simply less exposed to these risks, in our view.

As hyperscaler issuance continues to influence global credit markets, investors may find that sometimes what matters is not only what a market contains, but also what it does not.

 

* For illustrative purposes only. Reference to a particular security is on a historic basis and does not mean that the security is currently held or will be held within an L&G portfolio. The above information does not constitute a recommendation to buy or sell any security.

Assumptions, opinions, and estimates are provided for illustrative purposes only. There is no guarantee that any forecast will come to pass. Past performance is not a guide to the future. 

[1] Company disclosures, capital expenditure guidance and consensus estimates, August 2026.
[2] Mizuho: Reaching Saturation Point in Hyperscaler Supply? July 2026
[3] Alphabet Inc. bond prospectus / offering memorandum, February 2026.