Google plans a shared AI training preference: what does it mean for your website?
Google announced plans on 30 October 2026 to support a shared AI training preference in robots.txt. We explain what is proposed, what already exists and what business owners should do now.
A simpler choice about AI training
On 30 October 2026, Google announced plans to support a shared signal in robots.txt that would allow site owners to express preferences about the use of their content for AI training. Rather than maintaining separate instructions for each provider, publishers could communicate a broader preference to participating AI companies.
The crucial detail: this is a proposal, not a universal setting already available today. Google says it aims to support the mechanism in 2027, alongside other providers.
What does robots.txt do?
A robots.txt file typically lives at yourdomain.be/robots.txt. It tells automated crawlers which areas of a site they are allowed to request.
It is not a technical lock. Crawlers need to respect the instructions, and the file does not replace copyright law, privacy obligations or contractual terms about your content.
What is Google proposing?
The announcement builds on work by the IETF's AIPREF working group. A shared policy token in robots.txt would allow a website to express a broad AI-training preference.
Google says that when a publisher opts out through this future shared signal, it plans to interpret the choice as disallowing Google-Extended, provided no separate explicit Google-Extended setting overrides that interpretation. Site owners could still set provider-specific preferences.
That does not mean all AI providers will immediately honour the signal. Adoption, implementation and compliance will matter.
Will opting out of AI training remove your site from Google Search?
Not automatically. Google-Extended is a distinct control related to some uses of content for Google AI products. Ordinary web search crawling, indexing, search visibility and AI uses are different issues.
A business should therefore avoid blocking Googlebot or every crawler simply because it wants more control over AI training. Such a step could damage the visibility of an otherwise healthy website.
Five useful actions for businesses now
- Review your robots.txt. Check that important pages and assets are not unintentionally blocked and that your sitemap is discoverable.
- Define the outcome you want. Restricting AI training and remaining visible in search are separate objectives that require careful choices.
- Record existing bot rules. Know which providers you manage today and who has access to change the configuration.
- Avoid unverified policy tokens. Google's shared mechanism is planned for 2027. Do not paste experimental syntax into a live site just because someone claims it works.
- Monitor Google Search Console. If crawler settings change, watch for unexpected drops in discovery, indexing or traffic.
What should you do now?
Start by checking your existing robots.txt: which crawlers have access and why? Only change the rules once you know what you want to restrict. The shared signal is still being developed.
Source: Google Search Central, 30 October 2026. The planned implementation and timeline may change.