# AI CapEx

> **TL;DR.** AI CapEx is the industry shorthand for capital expenditure on AI infrastructure — the land, power, GPUs, servers, and data centers that hyperscalers are building to train and serve large models.

- **Category:** AI / Investing / Infrastructure
- **Stage:** established
- **Age:** 1000 days
- **Origin date:** 2023-09-20
- **First detected:** 2026-04-18
- **Canonical URL:** https://earlyterms.com/term/ai-capex
- **Sources:** 8 primary URLs

## Definition

AI CapEx is the industry shorthand for capital expenditure on AI infrastructure — the land, power, GPUs, servers, and data centers that hyperscalers are building to train and serve large models. It has become a top-line financial metric for Microsoft, Alphabet, Amazon, Meta, and Oracle, and the single biggest variable in AI-era stock analysis.

The term crystallized around [David Cahn's September 2023 Sequoia essay "AI's $200B Question"](https://sequoiacap.com/article/follow-the-gpus-perspective/) and Sequoia's [July 2024 "Game Theory of AI CapEx"](https://sequoiacap.com/article/ai-optimism-vs-ai-arms-race/), which called it "a euphemism for building physical data centers with land, power, steel and industrial capacity." By 2026 the top five hyperscalers committed [$660-690 billion for the year](https://futurumgroup.com/insights/ai-capex-2026-the-690b-infrastructure-sprint/).

## Why it's emerging now

Hyperscalers committed $660-690B for 2026, roughly doubling 2025 and consuming nearly 100% of operating cash flow vs a 10-year average of 40%. Circular OpenAI-Nvidia-Oracle deals, $121B in 2025 hyperscaler bond issuance, and Microsoft's power-constrained Azure backlog have pushed the 'AI CapEx' debate from finance desks into mainstream macro commentary.

## Related terms

- *parent:* hyperscaler
- *related:* AI bubble
- *related:* circular financing
- *related:* Stargate Project
- *related:* data center buildout
- *alias:* AI infrastructure
- *related:* GPU arms race
- *related:* sovereign AI
- *related:* intelligence-age
- *related:* ai-stack
- *alias:* AI's $200B Question

## Sources

1. [Sequoia — AI's $200B Question (David Cahn, origin framing)](https://sequoiacap.com/article/follow-the-gpus-perspective/)
2. [Sequoia — The Game Theory of AI CapEx](https://sequoiacap.com/article/ai-optimism-vs-ai-arms-race/)
3. [Paul Kedrosky — Honey, AI Capex Ate the Economy](https://paulkedrosky.com/honey-ai-capex-ate-the-economy/)
4. [CNBC — Tech AI spending approaches $700 billion in 2026](https://www.cnbc.com/2026/02/06/google-microsoft-meta-amazon-ai-cash.html)
5. [Futurum — AI CapEx 2026: The $690B Infrastructure Sprint](https://futurumgroup.com/insights/ai-capex-2026-the-690b-infrastructure-sprint/)
6. [Bloomberg — AI Circular Deals: How Microsoft, OpenAI and Nvidia Keep Paying Each Other](https://www.bloomberg.com/graphics/2026-ai-circular-deals/)
7. [Data Center Dynamics — US hyperscaler capex to top $700bn in 2026 (Moody's)](https://www.datacenterdynamics.com/en/news/us-hyperscaler-capex-to-top-700bn-in-2026-investors-fear-overbuild-and-weak-returns-moodys/)
8. [Goldman Sachs — Why AI Companies May Invest More than $500 Billion in 2026](https://www.goldmansachs.com/insights/articles/why-ai-companies-may-invest-more-than-500-billion-in-2026)

---
_Generated by EarlyTerms · https://earlyterms.com/term/ai-capex_