The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →To evaluate whether an AI infrastructure company can afford its expansion, look beyond its debt balance. Compare cash and liquid investments, operating cash flow, free cash flow after capital spending, interest costs, maturities, lease commitments and financing outside the parent company. Then ask whether planned capacity is built, powered, commissioned and earning revenue—and whether the company’s customers can pay. The same spending level can pose very different risks for a diversified cloud provider and a specialist operator dependent on a few projects or customers.
Start with the full financing picture, not a single debt ratio
Debt-to-EBITDA or net debt alone cannot show whether an infrastructure buildout is financeable. A company can carry substantial debt while generating dependable cash from established businesses; another can have less reported debt but face pressure from leases, project-level borrowing, weak liquidity or delayed revenue. No universal safe debt-to-EBITDA threshold is established for this sector.
For each company, align the reporting period and examine these measures together:
- Debt and leases: gross debt, net debt after cash, lease liabilities, secured or project-level borrowing, debt maturities and any financing through special-purpose vehicles (SPVs).
- Cash available to service obligations: cash and liquid investments, operating cash flow, interest expense, interest coverage and access to additional liquidity.
- Cash left after investment: free cash flow after capital expenditures (capex). Rapid expansion can reduce or turn this measure negative even when operating cash flow is strong.
- Investment intensity: capex relative to revenue and operating cash flow. Separate amounts already spent from forecasts, commitments and announced plans.
- Funding flexibility: assess the likely cost and trade-offs of bonds, private credit, equity issuance, leases and project finance rather than treating them as interchangeable.
Include both parent-company obligations and financing tied to individual projects. The OECD notes that corporate bond totals do not capture all SPV issuance, so a headline debt figure may not represent the full financing exposure.
#1 Best Overall
Check whether spending is turning into usable, revenue-producing capacity
A facility announcement or completed building is not the same as productive capacity. A project may still need grid access, transmission, equipment, power energization, commissioning and customer workloads before it generates the cash expected to service its financing. Track each stage and the time between investment and revenue, rather than counting announced capacity as though it were already operating.
Power constraints matter to credit assessment because a site that is built but not energized or operational may not produce cash flow on schedule. Moody’s also points to tenant credit quality and contract protections as relevant monitoring factors. A long-term lease can support expected revenue, but it does not by itself remove construction, power-delivery, commissioning or utilization risk.
The scale of the constraint could grow: the International Energy Agency projected global data-center electricity consumption of 485 TWh in 2025, rising toward approximately 950 TWh in 2030, as reported by Moody’s in 2026. These are projections, not measurements of a particular company’s available power or utilization.
Evaluate the customer and business model behind the financing
Infrastructure is financeable only if demand becomes collectible revenue. Examine customer concentration, contract length, counterparty creditworthiness, termination terms and any reliance on interconnected suppliers, investors or customers. A contract can reduce demand uncertainty without guaranteeing timely project completion or actual utilization.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Business-model differences are essential to comparison. A diversified hyperscaler may use cash from established businesses beyond AI infrastructure to support investment. A specialized operator may depend much more directly on a small customer base, project completion and high utilization. A similar debt balance therefore does not imply a similar ability to absorb delays or customer losses.
Use current figures as dated examples, not a sector-wide verdict
Recent figures show the scale of investment and financing without proving that all companies face the same risk:
| Figure | What it measures—and what it does not |
|---|---|
| $122 billion of hyperscaler corporate bond issuance in 2025 | The OECD’s March 2026 report says this was 45% of global technology-firm bond issuance and the largest amount in real terms in its series. It covers bonds issued directly by companies and may omit SPV financing; it is not a measure of total sector debt or distress. |
| $4.1 trillion of cumulative hyperscaler capex in 2026–2030 | An OECD consensus estimate, not realized spending. Its period and company scope differ from the IMF estimate below. |
| $3.4 trillion of AI-related capex through 2029 | An IMF estimate from its April 2026 Global Financial Stability Report. It is not directly comparable with the OECD’s differently scoped 2026–2030 estimate. |
| Oracle: $55.7 billion of capex in FY2026 versus $21.2 billion in FY2025 | Oracle’s Form 10-K for the year ended May 31, 2026 reports these company-specific figures and attributes the increase primarily to data-center expansion. They are not sector averages. |
| Oracle: $45–50 billion of planned gross funding proceeds in calendar 2026 | Oracle’s February 1, 2026 announcement described a plan to raise proceeds through a combination of debt and equity to expand OCI capacity for contracted demand. The plan is not evidence that all the proceeds were raised. |
Oracle said it was raising money “in order to build additional capacity to meet the contracted demand from our largest Oracle Cloud Infrastructure customers, including AMD, Meta, NVIDIA, OpenAI, TikTok, xAI and others.” That statement gives the company’s stated rationale for the financing plan; it does not establish when the capacity will be delivered, how much will be used or whether every planned dollar was raised.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Balance today’s strength against future funding needs
The OECD’s March 2026 Global Debt Report describes rising capex intensity and lower free-cash-flow ratios for many hyperscalers in 2025, with leverage increasing in some cases. It also says leverage broadly remained manageable, given historically low debt funding. The IMF’s April 2026 report similarly describes major hyperscalers as having strong balance sheets and free cash flow while warning that estimated AI-related investment could pressure balance sheets in future years. These findings argue against both extremes: high spending alone does not establish distress, and a strong current balance sheet does not guarantee that future investment will remain easy to fund.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteBest Value
Financing links can also transmit stress. The IMF warns that circular financing—where companies in the AI value chain finance one another or depend on interconnected arrangements—can amplify adverse shocks. When reviewing a company, consider whether expected demand, capital availability and counterparties are independent enough to withstand a setback elsewhere in that chain.
Keep estimates, plans and actuals distinct. The OECD’s projected $4.1 trillion of cumulative capex for 2026–2030 and the IMF’s $3.4 trillion estimate through 2029 have different scopes and periods; neither is realized spending. Likewise, a company’s announced funding target should be checked against subsequent filings before treating it as completed financing.
Quick Recap
A practical company-by-company review
- Align the period and scope. Use filings for comparable fiscal or calendar periods, and separate the parent company from project-level entities where possible.
- Map all obligations. Record debt, leases, secured borrowing, maturities and disclosed SPV exposure—not just a headline net-debt figure.
- Test the cash bridge. Compare liquidity and operating cash flow with interest, scheduled repayments and capex. Note whether free cash flow after capex is improving or being consumed by expansion.
- Trace capacity to revenue. For major projects, check construction, power access, energization, commissioning, utilization and customer revenue rather than relying on announced capacity.
- Assess demand quality. Review concentration, contract duration, counterparty credit and the degree to which projected cash flow depends on a few customers or connected firms.
- Examine how the next expansion is funded. Distinguish issued debt and raised equity from plans, commitments and forecasts, then consider the cost and flexibility of the funding mix.
- Revisit the conclusion as projects progress. New borrowing, a delayed energization date, weaker utilization or a change in customer credit can alter the cash-flow outlook even if reported debt has not changed.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
