Hyperscaler capex to AI revenue (trailing year)
Trailing four quarters of filed capex at Microsoft, Alphabet, Amazon, Meta and Oracle divided by AI revenue, taken as the frontier labs' latest run-rates plus Menlo's enterprise generative-AI spend. Derived metric capex_to_revenue_stack.
Reason through thisThe reading
Hyperscaler capex is not all AI and the revenue side mixes annualised run-rates with one annual survey, so read the direction, not the level; grade C by construction.
8 readings drawn from the derived metric capex_to_revenue_stack; 98 observations in the evidence.
Derived (capex_to_revenue_stack) from: SEC EDGAR XBRL company facts; Epoch AI data hub — AI companies; Menlo Ventures, State of Generative AI in the Enterprise. SEC EDGAR, XBRL company facts API. Epoch AI, 'Data on AI Companies' (CC BY 4.0), epoch.ai/data. Menlo Ventures, '2025: The State of Generative AI in the Enterprise'.
Why it matters
Sequoia's $600B question as a ratio: how far the buildout runs ahead of the revenue that has to repay it. A wide gap can mean the spending is early and revenue will follow, or that the builders will never earn it back; the ratio cannot say which, only whether the gap is closing.
What cuts against it
Both sides are approximations: capex includes non-AI spend and the revenue side double-counts enterprise purchases from labs; a ratio of filed capex to press-reported run-rates is the brief's own 'gap #1' caveat. Part of the ratio's fall is a step when Menlo's 2025 survey wave raised the enterprise-spend denominator, not a change in capex.
How it is measured
- derived metric
capex_to_revenue_stack(formula in the semantic layer) - source Epoch AI data hub — AI companies · default tier 5 · CC BY 4.0
- source Menlo Ventures, State of Generative AI in the Enterprise · default tier 6 · Report
- source SEC EDGAR XBRL company facts · default tier 4 · Public domain (U.S. government); fair-access policy requires a descriptive User-Agent
- proxy capital_flow · cadence quarterly
Four quarters; a move under 0.3x is run-rate noise. Rising = capex pulling further ahead of revenue (the bet growing); falling = revenue catching the buildout.
Applied to metric:capex_to_revenue_stack.
Derived rows (8)
Latest readings
Why this status, and its history
Evaluator: 2.507 (SEC EDGAR XBRL company facts, 2026-06-30) reads dispersing over 4 periods with a dead band of 0.3. Auto-reason; rule rationale: Four quarters; a move under 0.3x is run-rate noise. Rising = capex pulling further ahead of revenue (the bet growing); falling = revenue catching the buildout.
- 2026-09-10unmeasured to dispersingconf — → 40 · the evaluator (automatic)
Evaluator: 2.507 (SEC EDGAR XBRL company facts, 2026-06-30) reads dispersing over 4 periods with a dead band of 0.3. Auto-reason; rule rationale: Four quarters; a move under 0.3x is run-rate noise. Rising = capex pulling further ahead of revenue (the bet growing); falling = revenue catching the buildout.
Why this confidence
Rests on 98 observations from Epoch AI data hub — AI companies, Menlo Ventures, State of Generative AI in the Enterprise, SEC EDGAR XBRL company facts; best evidence tier 4, grade A; current.
Confidence is independent of status: 90–95 multiple strong independent sources; 70–89 good evidence, some ambiguity; 50–69 mixed or hard to operationalise; 0–49 limited or vague evidence.
Predictions that lean on this
- Happening
- Happening
- Both sides expect this
- Too early to tell
- Too early to tellAI's revenue falls far short of what the spending on it needs to repay. · David Cahn (Sequoia Capital)