Glossary · 3 · Working with models: prompts and context

Structured output

Also known as: JSON mode, Schema-constrained output

Structured output is model output that follows a predefined, machine-readable format — typically JSON matching a schema — so that other software can process it reliably. Many AI APIs can enforce a schema during generation.

Intermediate
Expert
Technical writers
Developers

In one sentence

Structured output explained: making AI return JSON or XML that follows a schema — the bridge between AI and content pipelines.

Example

A pipeline asks a model to extract product name, variant, hazard and required action from legacy manuals and return them as JSON matching a schema, which then feeds a DITA conversion.

Why it matters on your learning path

Technical writers: Structured output connects AI to structured authoring: metadata extraction, tagging and conversion to DITA or iiRDS metadata.
Developers: Use schema enforcement where the API offers it, and validate anyway before writing to downstream systems.

Related to

Schemas and JSON-LD in the information architecture glossary describe the same idea from the content-modeling side.