Amazon’s £4.25 Billion Sterling Bond Debut Shows AI Funding Going Global

Amazon’s first sterling-denominated bond sale raised £4.25 billion ($5.76 billion), exceeding the initially expected £4 billion after drawing more than £12 billion of investor orders. The transaction gave Amazon a new currency market for funding its artificial-intelligence infrastructure buildout—and offered a fresh test of investor appetite for the widening wave of hyperscaler debt.17

The deal: four maturities and a £250 million upsizing

Amazon sold four sterling tranches maturing in 3, 6, 12 and 19 years. The final structure was:

• £1.25 billion of 3-year notes
• £1 billion each of 6-year, 12-year and 19-year notes

That brought total proceeds to £4.25 billion, £250 million more than the £4 billion initially expected. Orders exceeding £12 billion represented almost three times the original target, allowing the company to increase the short-dated tranche.17

JPMorgan, Barclays, HSBC and NatWest managed the offering.3

Final pricing was tighter than the initial guidance

The terms improved as investor orders accumulated. Initial discussions were around 70 basis points over UK government bonds for the shortest maturity and 110 basis points for the longest.3

Reported final pricing tightened to 53 basis points over gilts for the 3-year notes and 93 basis points for the longest tranche. In bond-market terms, lower spreads mean Amazon was able to borrow more cheaply relative to the UK government benchmark than initial indications suggested.

The exact final spread for each middle tranche is not established by the provided reporting. But the movement at both ends of the maturity range, alongside the upsized deal, is a clear indication that demand was strong.

Why a sterling debut matters

This was Amazon’s first bond sale in pounds, adding UK institutional investors and the sterling market to the company’s funding toolkit. That matters because the spending being financed—data centres, computing capacity and related AI infrastructure—is enormous and long-lived. Borrowing across several currencies can broaden the potential buyer base rather than relying solely on the US-dollar market.

The 19-year tranche is especially notable in that context: it extends Amazon’s funding well beyond the near term. Long maturities can help match long-lived infrastructure investment with long-dated financing, though they also expose investors to greater interest-rate sensitivity if government yields rise.

One leg of Amazon’s 2026 multi-currency borrowing campaign

The sterling transaction follows a series of major Amazon debt deals across markets:

• In March, Amazon raised €14.5 billion in an eight-part euro deal, described by LSEG as the largest euro corporate-bond transaction on record.21
• In June, it issued C$14 billion of Canadian-dollar notes in a five-part deal, a record for Canada’s corporate bond market.
• In July, Amazon sought $25 billion in an eight-part US-dollar offering to fund AI investment. Demand reportedly peaked at $62 billion before ending around $41 billion after terms tightened.20
• Reuters reported in August that Amazon was also preparing a first six-part Swiss-franc offering.22

The pattern is diversification: Amazon is tapping multiple investor pools and currencies as AI infrastructure spending increases its funding needs.

The contrast with Amazon’s US-dollar deal

The sterling sale’s robust order book contrasts with a more cautious signal from Amazon’s July dollar offering. Although the US deal was large and its peak orders reached $62 billion, the final book of about $41 billion equated to roughly 1.6 times the $25 billion deal size.20

That does not mean the US transaction failed—it was completed at substantial scale. But it illustrates an important distinction: investors may still be willing to fund high-quality technology borrowers while becoming more selective about price, maturity and the cumulative volume of new issuance.

Alphabet had already set the sterling benchmark

Amazon was not the first hyperscaler to use sterling debt for AI-era funding. Alphabet raised £5.5 billion in a five-part sterling sale in February as part of a $31.51 billion global bond raise. The deal included a rare 100-year bond, which raised £1 billion and carried a 6.125% coupon.

Alphabet’s transaction showed that long-dated sterling demand exists for the largest technology issuers. Amazon’s deal was less extreme in maturity, topping out at 19 years, but its successful debut confirms that sterling is becoming an important funding channel for US hyperscalers.

The bigger issue: can bond markets absorb the AI buildout?

Amazon’s strong sterling reception comes amid a much larger debt surge. Alphabet, Amazon, Meta, Microsoft and Oracle had issued about $220 billion of debt by early September 2026, more than double the previous year’s total at the comparable point, according to LSEG data cited by Reuters.

That pace has raised concerns beyond any one issuer. Reuters reported that major technology companies have faced higher borrowing costs as investors become more selective about absorbing the growing supply of AI-related bonds. The European Central Bank has also warned that heavy issuance by US tech groups in euro markets could crowd out other borrowers, raise financing costs—including for governments—and increase credit risks.

For investors, long maturities add another layer of risk: bond prices generally fall when benchmark yields rise, and longer-dated bonds are typically more sensitive to that move than short-dated bonds.

What Amazon’s sale signals

Amazon’s £4.25 billion debut shows that it can still broaden its funding base, upsize a transaction and tighten pricing when demand is strong. The sale is therefore a successful entry into the sterling market.

Its wider significance is less straightforward. The same AI investment cycle that supports demand for high-grade technology bonds is also creating an unusually large supply of them. Amazon’s deal suggests capital remains available; the next question is what yield concessions investors will require as more long-dated hyperscaler debt arrives.