> ## Content Index
> Fetch the complete content index at: https://paulkedrosky.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# How Much Revenue Must AI Deliver?
- URL: https://paulkedrosky.com/how-much-revenue-must-ai-deliver/
- Published: 2026-10-23T20:39:22.000Z
- Updated: 2026-10-23T20:39:22.000Z
- Description: A new interactive tool for testing the revenue required to justify AI infrastructure spending.
- Author: Paul Kedrosky

AI’s infrastructure cost is now **measured in trillions**. The harder question is what **revenue that spending implies**, because the math is backwards, with revenue uncertain, but **capex increasingly locked in**.

Various groups have **estimates**, like researchers at [Columbia](https://www.brookings.edu/wp-content/uploads/2026/09/4c%5FVan-Nieuwerburgh.pdf) and analysts at [Bain](https://www.bain.com/insights/new-innovation-is-required-to-fund-ais-6-trillion-buildout-technology-report-2026/), but those are static. We wanted something that allowed us to **play "what if" with existing forecasts**, or create our own.

We have built a new interactive tool for premium subscribers, [**How Much Revenue Must AI Deliver?**](https://paulkedrosky.com/tools/revenue-hurdle/), to make that arithmetic less opaque. **Start with the capital outlay**, choose a **return** hurdle and **cash margin**, and see the **implied annual revenue** the investment needs to support it. The chart shows how that answer changes as spending rises.

![Top of the AI revenue calculator showing the $3.7 trillion annual requirement and comparisons with software revenue, U.S. GDP, and white-collar labor.](https://paulkedrosky.com/content/images/2026/10/clipboard-1.png)

See how required revenue compares with software revenue, GDP, and white-collar labor.

You can start from published Columbia or Bain estimates, or choose Custom and enter your own aggregate or annual capex. Then change the **assumptions**. A larger buildout, a higher required return, slower depreciation, or a thinner margin can dramatically move the revenue target. And it is **all put in context** against other large markets, like global software, or, you know, U.S. GDP.

## Video walkthrough

You can **t**[**ry the tool here**](https://paulkedrosky.com/tools/revenue-hurdle/). It is available to all **premium subscribers**.