The Explain API

The Explain API lets a program create explainers the way you create them by hand today: you send a topic and the material to teach from, and you get back an explainer with narration, slides, and its own URL.

It isn't released yet. If you want it, leave your email and we'll tell you when it opens.

One email when it launches. Nothing else.

Use cases

Anywhere someone has to understand something is a place an explainer can go, so the list is long. A few examples:

Batch generation across languages. One topic, 33 languages, in a single run. A glossary, a course syllabus, or a back catalogue of articles becomes a set of explainers without anyone sitting in the composer repeating themselves.

Inside your own product. Generate an explainer when your user needs one, then embed it in your app: an error message that explains itself, a chart with a lesson behind it, an onboarding flow that teaches instead of pointing at a tooltip.

Learning content at scale. Course platforms, internal training, and documentation teams producing lessons per chapter rather than per quarter. This docs site already puts an explainer on Python chapters; the API is how you do that for a thousand pages.

Until then

Two things already work programmatically. Coding agents create explainers through the MCP server, with your repository as the source material, and ChatGPT and Codex do the same through the plugin. Both write the lesson themselves, which is the part the API would hand back to you.