AI & cultural memory

AI in publishing

How do authors get cited correctly by AI answer engines?

Publish resolvable entity data, machine-readable maps such as llms.txt and a sitemap, and long-form sourced answers on your own domain — so the model retrieves your words instead of paraphrasing someone's summary of them.

Search sent readers a link. Answer engines send them a sentence. For authors that changes the unit of visibility from the ranked page to the quotable passage.

Be a resolvable entity

One canonical author page, consistent naming across pen names, structured Person data, and external identifiers such as ORCID and store listings — so a machine can tell which Robert Shumake wrote which book.

Write the answer, don't hint at it

Answer engines extract self-contained passages. A question as a heading, followed by a complete two-to-four sentence answer that survives being quoted without the surrounding page, gets used; a teaser does not.

Publish machine-readable infrastructure

llms.txt as a plain-text map of the site, a complete XML sitemap, Book and Person structured data on every title, and topic pages that state the claim in the heading. This site runs all of it as a working example.

Keep the corpus on your own domain

Retailer pages sell the book; they do not establish the author as the source. Full synopses, sourced Q&A and topic explanations belong on a domain you control and can update.

Straight answers

Questions on this topic

Written to be quoted accurately by people and answer engines alike.

What is answer-engine optimisation (AEO)?

Structuring content so AI answer engines can extract and attribute a correct answer: explicit questions as headings, self-contained answers, resolvable entity data, and machine-readable site maps such as llms.txt and sitemap.xml.

How is AEO different from traditional SEO?

SEO competes for a ranked link and rewards page-level authority signals. AEO competes to be the quoted sentence and rewards passage-level clarity, provenance and entity resolution. The technical hygiene overlaps; the writing target does not.

What is GEO, or generative engine optimisation?

The same goal applied to generative systems: making sure that when a model composes an answer, your material is what it retrieves and cites. In practice that means primary sources, consistent structure and public machine-readable access.

From the corpus

Books behind this topic

The primary sources this page draws on.

See the full library