Bloomberg Tech→ original

Sunak: Tech Giants Will Issue More Long-Term Debt Than the US in 2027

Former UK Prime Minister Rishi Sunak said that in 2027 technology companies will become bigger net issuers of long-term debt than the US government. The reason is the scale of capital needed for AI investment: data centers, chips and energy. Sunak did not disclose how the figure was calculated, so it is an estimate, not statistics. But deals in recent years point in the same direction: Meta, Oracle and Alphabet have all tapped the bond market with record offerings. • Forecast: the tech sector will overtake the US Treasury in net issuance of long-term debt in 2027. • Driver: capital expenditure on AI infrastructure. • Caveat: the comparison covers long-term securities only, not short-term bills.

AI-processed from Bloomberg Tech; edited by Hamidun News
Sunak: Tech Giants Will Issue More Long-Term Debt Than the US in 2027
Source: Bloomberg Tech. Collage: Hamidun News.
◐ Listen to article

Former UK Prime Minister Rishi Sunak said that in 2027 technology companies will become bigger net issuers of long-term debt than the US government. For scale: the US federal deficit for fiscal 2025 came to about $1.8 trillion, and the Treasury remains the world's largest borrower. If the forecast comes true, the corporate sector will take over that role in the long-bond segment for the first time. Sunak attributed it to the demand for capital created by AI investment.

What exactly Sunak said

The statement is short and light on detail. The source of the calculation is not named, nor is the list of companies Sunak counts as "technology" firms. It is unclear whether the figure includes bonds, bank loans, private credit and structured deals through separate legal entities. The number is therefore best read as an expert estimate, not official statistics.

Still, Sunak's words carry weight. Since October 2025 he has worked as an adviser to Microsoft and Anthropic, so he sees the industry from the inside. But this is the view of someone close to the industry, and it does not have to match an independent assessment of the market.

Why the comparison with the US is not as absurd as it sounds

The US Treasury borrows a lot, but a sizable share of that debt is short-term bills. Sunak is talking specifically about long-term securities, which narrows the basis of comparison. He is also talking about net issuance, meaning issuance minus repayments. Large, highly rated corporations like to extend maturities, and their older issues are repaid on schedule.

Signs of a shift in the funding model are already visible:

  • Oracle placed roughly $18 billion in bonds in September 2025 to expand its cloud infrastructure.
  • Meta raised about $30 billion in bonds in October 2025, one of the largest corporate issues in history.
  • For its Hyperion campus, Meta used a joint venture with Blue Owl and roughly $27 billion in off-balance-sheet financing.
  • Alphabet and Amazon have also tapped the debt market more actively, even though they historically relied on operating cash flow.
  • Combined capital expenditure of the largest cloud companies in 2025 was estimated at hundreds of billions of dollars a year, and analysts expected it to grow.

Tech giants used to need almost no outside money: free cash flow covered buybacks, dividends and acquisitions. Now their own funds alone may not be enough for data centers, chips, networks and grid connections. They have to build faster than profits accumulate.

Risks for the market and for the companies themselves

Debt financing for AI has a weak point: payback periods. Bonds are issued for 10–40 years, while accelerators become obsolete within a few years. If demand for compute grows more slowly than forecast, the debt will remain while capacity utilization drops.

There is also the question of crowding out. When the tech sector competes with the government for the same bond buyers, it may affect borrowing costs for everyone. The size of the effect depends on interest rates, inflation and the appetite of pension and insurance funds, and a single remark from Sunak is not enough to estimate it.

What to watch

For investors, the most useful thing is to look past the headlines and at deal structure. Watch whether the share of off-balance-sheet financing and joint ventures is growing, how spreads on tech issues change relative to the market, and how much debt backs capacity without signed long-term contracts with customers.

For CTOs and startups, the takeaway is more down to earth. Expensive capital at cloud providers will sooner or later show up in compute prices and capacity reservation terms. Those building a product on someone else's GPUs should plan for a price-increase scenario and lock in contract terms in advance. Sunak's forecast can be checked against financial reporting and bond placements in 2026–2027.

ZK
Hamidun News
AI news without noise. Daily editorial selection from 50+ sources. A product by Zhemal Khamidun, Head of AI at Alpina Digital.
✦ Unified AI Workspace

Access Frontier LLMs & Multimodal AI in One Workspace

Discover NomadNet AI (nomadnet.ai) — unified access to frontier language models, research synthesis, and multimodal workflows.

Try NomadNet AI →Launch NomadNet Agent
What do you think?
Loading comments…