Does Google Read Aloud respect robots.txt?

Only partially. A rule for `Google-Read-Aloud` may be applied, but a user-initiated request can still reach a URL that the rule disallows.

Does Google Read Aloud crawl links across my site?

The documented workflow fetches the page selected by a user. The record does not describe site-wide discovery, indexing, or a recurring crawl schedule.

What will users experience if I block it?

Google's service may be unable to retrieve the page and read it aloud. That mainly affects people using the text-to-speech function for the blocked URL.

Is this request evidence of AI search or model training?

No. The documented purpose is text-to-speech for a requested page. A visit does not establish an AI citation, search inclusion, or use in model training.

Google Read Aloud AI Crawler Guide

What is Google Read Aloud?

Google Read Aloud retrieves a page when a person asks Google's service to read it aloud. The service converts the fetched page into speech with text-to-speech, so the request is tied to that user's chosen URL rather than a background attempt to crawl the whole site.

A Google-Read-Aloud log entry therefore records an accessibility fetch. The service needs the page at request time, and blocking the request can prevent that user from hearing the content. The record does not describe link discovery or scheduled recrawling.

Robots.txt support is partial because the fetch happens on a user's instruction. A bot-specific disallow can express the site's preference, but it may not stop every request. Authentication and server access rules remain the controls for material that must not be returned to an automated client.

Read Aloud is not a search indexing bot, and a fetch does not indicate that Google selected the page for an AI answer. The source material also makes no claim that the page is added to a training data set. Its documented use is limited to reading the requested web page with text-to-speech.

Relevant for AI search

Is Google Read Aloud relevant for AI search?

Yes. Google Read Aloud collects pages for an AI product, so what it can crawl influences how AI systems describe your brand.

Google Read Aloud fetches, indexes, or retrieves pages for an AI product, so the pages it can reach shape how that system describes your brand and products.

How to handle Google Read Aloud

Keep Google Read Aloud available for public pages when you want users to access them through Google's text-to-speech feature. Protect private pages with authentication instead of assuming a crawler directive makes them private.

The corresponding record shows Google-Read-Aloud in the HTTP user agent. A valid preference for that token is:

User-agent: Google-Read-Aloud
Disallow: /

Compliance is only partial. Because an individual user initiates the fetch, Google Read Aloud may still request a disallowed URL. Check logs after publishing the rule, and enforce any necessary denial at the server or edge with the accessibility tradeoff in mind.

Examples

A commuter asks Google to read one long article aloud. The article server sees a `Google-Read-Aloud` request for that URL, not a sequence covering the surrounding section.
A publisher disallows the fetcher in robots.txt and still records an occasional request initiated by a reader. It moves sensitive material behind authentication instead of treating the robots rule as an access boundary.

Frequently asked questions about Google Read Aloud

Learn about AI visibility monitoring and how Promptwatch helps your brand succeed in AI search.

A person asks the Google Read Aloud service to read a particular web page. The fetch is made so the service can obtain that page for text-to-speech output.

Related bots

meta-externalfetcher

The Meta-ExternalFetcher crawler performs user-initiated fetches of individual links to support specific product functions.

User Initiated

Google Read Aloud