In a Nov. 10 announcement, WordPress.com said AI crawler controls and other SEO tools are rolling out to Premium, Business and Commerce sites under Jetpack → SEO.
The update is for WordPress.com, Automattic’s hosted WordPress service. It should not be read as a universal change to every self-hosted WordPress installation running the open-source WordPress software.
The new GEO tab, short for generative engine optimization, separates supported AI crawlers into answer engines and training crawlers. WordPress.com’s support guide says answer engines help AI tools answer questions and cite sources, while training crawlers collect content for model training. On eligible plans, site owners can mark individual crawlers as disallowed or use group-level controls.
Those choices are written to the site’s robots.txt file. WordPress.com describes robots.txt as the open protocol sites use to tell crawlers what they can access, but it also says each AI platform decides whether to follow the file. That makes the setting a machine-readable request, not an enforcement layer.
AI crawler access is allowed by default, according to WordPress.com support documentation. The settings take effect when a site is public and visible to search engines; private, Coming Soon and temporary staging sites see a notice instead of the AI crawler controls.
There is still a site-wide opt-out on every plan
WordPress.com also points to a broader Prevent third-party sharing setting under Settings → Reading → Site Visibility. That setting is available for every WordPress.com site and is designed to exclude public content from WordPress.com’s network of third-party content and research partners, including AI platforms that use content for model training.
When enabled, WordPress.com says the setting also adds known AI bots to the site’s robots.txt disallow list. The same caveat applies: AI platforms still choose whether to honor the request.
This setting matters because it can override the newer GEO tab. WordPress.com says Prevent third-party sharing takes priority over GEO tab choices for the crawlers it covers, and the GEO controls cannot be changed for those crawlers until the broader setting is turned off.
The answer-engine split is the useful part
Separating AI search visibility from AI training is becoming standard across major AI and search providers. OpenAI’s crawler documentation separates OAI-SearchBot, which is used for ChatGPT search results, from GPTBot, which crawls content that may be used to train OpenAI’s generative AI foundation models. Google documents Google-Extended as a control for future Gemini model training and certain grounding uses; Google also says that Google-Extended does not affect a site’s inclusion or ranking in Google Search.
For publishers, that split creates a real tradeoff. Blocking the wrong crawler can reduce a site’s chance of being cited or surfaced in AI answers, while allowing every crawler may leave content available for uses the publisher does not want. WordPress.com’s grouping makes that choice easier to see, even if it does not remove the tradeoff.
llms.txt support has limits
The same GEO tab can generate an llms.txt file, which WordPress.com describes as an optional summary of a site for AI tools. The file is published at the site’s domain followed by /llms.txt.
That should not be treated as a universal AI discovery standard. WordPress.com’s support guide says not every AI tool reads llms.txt and Google Search ignores it. In practical terms, llms.txt can sit alongside sitemaps, schema and clear on-page structure, but it does not replace them.
For broader context, Tech Help Canada has separately covered how clearer structure, useful headings and accurate structured data can make content easier for AI search systems to understand.
The SEO dashboard is broader than AI crawlers
Beyond GEO, the new WordPress.com SEO page centralizes several controls that had been spread across screens. The Overview tab shows whether search engines can see the site, whether it is verified, whether AI crawlers can reach it and how many posts or pages still need an SEO title or meta description.
The Content tab lets publishers review and edit SEO titles and descriptions without opening each post. The Settings tab includes title formats, front-page descriptions, site verification, structured data, author profile details, canonical URL settings, social previews and sitemap options. WordPress.com support also lists AI-generated SEO metadata and image alt text as available from the SEO tools.
WordPress.com says the tools are built into the platform, so a separate SEO plugin is not required. Paid-plan users can still add an SEO plugin if preferred.
The timing lines up with wider AI preference work
The WordPress.com launch lands less than two weeks after several AI and search companies backed a shared robots.txt signal for AI training preferences.
On Oct. 30, OpenAI said it would honor a common robots.txt signal that lets website operators express a generative AI training preference across multiple participating providers, with implementation planned for the coming months. Google said it plans to support a shared AI training control in 2027, and its crawler documentation says support for the AIPREF signal is intended in early 2027. Microsoft said its new AI Training and AI Grounding preferences will begin taking effect as pages are crawled starting Feb. 1, 2027.
The underlying IETF Internet-Draft defines a __usage_ai-training policy token, but the draft itself says it has no formal standing in the IETF standards process. WordPress.com’s announcement did not say its new dashboard supports that shared token. The current rollout focuses on WordPress.com’s own per-crawler controls, llms.txt generation and SEO settings.
What site operators should check now
- Plan access: Full SEO and GEO tools are rolling out to Premium, Business and Commerce plans. Free and Personal sites get site visibility and search engine verification controls from the same SEO area.
- Current visibility: AI crawler controls depend on the site being public and visible to search engines.
- Third-party sharing status: The Prevent third-party sharing setting can override GEO tab choices for covered crawlers.
- Search versus training: Answer-engine access and training access should be reviewed separately, since blocking one use can have different visibility consequences than blocking another.
- Robots.txt after changes: The final public
robots.txtfile should be checked after settings are changed, especially on sites using other SEO tools or custom crawler rules.
The practical value of the update is accessibility. Many small publishers do not edit robots.txt directly, and many do not know which AI crawler names map to search, training or user-requested fetching. Putting those choices next to metadata, schema and verification makes AI access part of ordinary site management.
Related: Tech Help Canada’s WordPress review compares the SEO tools and site management options in hosted WordPress.com and self-hosted WordPress.
I can suggest related Tech Help Canada articles based on the topic you’re reading now.
Tech Help Canada Staff researches, writes, and reviews practical content for business owners and professionals. Our coverage spans business, marketing, SEO, technology, and the tools and systems people use to grow and operate online. We focus on clear, useful information backed by research, hands-on experience, and editorial review. Learn more about our team and editorial standards. Need help with something? Contact Us
