Reflexa Technologies
Reflexa Technologies
Yes, for search-backed answers. A noindex page is left out of the same search indexes that ChatGPT's search mode, Perplexity and Google's AI Overviews retrieve from, so there's nothing there for any of them to quote.
This is the access layer the what is GEO pillar puts first in the chain: access, then identity, then content, then trust. A noindex tag fails that first layer just as completely as a blocked crawler, it just does it quietly, inside a setting most people never open after launch day.
Because noindex is an instruction to leave a page out of a search index, and ChatGPT's search mode, Perplexity and Google's AI Overviews all answer by retrieving from a search index rather than by asking the live page a question in real time. Google states directly that AI Overviews and AI Mode are built on the same indexing and ranking systems as regular Search, with no separate "AI index" a page could opt into instead. A page that has told Search not to index it has, by the same stroke, told the layer AI Overviews reads from not to index it either.
Proof: Google — AI features and your website · Google — Robots meta tag and X-Robots-Tag specifications.
Why it matters: this isn't a Google quirk you can avoid by targeting a different engine. Perplexity and OAI-SearchBot both build their own index ahead of answering, the same shape of system noindex was built to opt out of.
No, the two controls sit at different points in the same pipeline, and mixing them up is how a noindex tag goes unnoticed for months. robots.txt decides whether a crawler is allowed to fetch the page at all. noindex assumes the fetch already happened, then tells the search system not to keep what it found.
| Control | What it governs | What a crawler sees |
|---|---|---|
| robots.txt Disallow | Whether the page can be fetched at all | Never reaches the page; nothing to read |
| <meta name="robots" content="noindex"> | Whether a fetched page is added to the index | Reads the page, then discards it from the index |
| X-Robots-Tag: noindex (HTTP header) | Same as the meta tag, set at the server or CDN instead of in HTML | Reads the page; the instruction never appears in the visible source |
Proof: Google's specification states these settings "can be read and followed only if crawlers are allowed to access the pages," meaning a robots.txt block can also hide a noindex rule from ever being seen: Google — Robots meta tag and X-Robots-Tag specifications.
Do this: checking robots.txt alone is not enough. A page can pass every robots.txt test and still be invisible in AI answers because of a noindex rule robots.txt never mentions.
Three places account for almost every accidental noindex: a plugin setting, a site-wide toggle, and a staging configuration that never got turned off. None of them require anyone to type "noindex" on purpose.
Why it matters: the header case is the dangerous one, since it leaves no trace in the page source a developer would normally glance at before shipping.
View the page source for the meta tag, then check the response headers separately, because one check never catches both. A browser's normal page view shows neither.
Do this: also run What AI Sees on the same page. If a page is readable there but still absent from AI answers, noindex and a handful of other access checks are the first place to look.
None of this requires buying anything, and each step takes a few minutes:
Bottom line: this is usually a one-click fix once the setting is found, the hard part is knowing where to look.
For search-backed answers, yes. A noindex page is left out of the indexes that ChatGPT's search mode, Perplexity and Google's AI Overviews retrieve from, so none of them have the page available to quote, the same way it disappears from ordinary Google results.
robots.txt controls whether a crawler can fetch the page at all. noindex assumes the page was fetched, then tells the search system not to add it to the index. A page can be fully fetchable and still invisible in answers because of a noindex tag or header.
Three common places: an SEO plugin's per-page "discourage search engines" box left on from a redesign, a site-wide WordPress setting under Settings, Reading, or an X-Robots-Tag header set at the server or CDN level for staging that shipped to production unchanged.
Google's own documentation confirms it removes noindex pages from Search and that AI Overviews and AI Mode build on that same index. Whether every AI crawler's own indexing pipeline treats the tag identically isn't spelled out the same way in each engine's documentation, so treat it as the industry-standard signal it's designed to be, and verify with the free check rather than assume.
View the page source and search for meta name robots content noindex, then check the response headers for X-Robots-Tag, since a header-based noindex never shows up in the HTML. The free noindex check runs both across your priority pages in one pass.
The free check reads your priority pages for both the meta tag and the X-Robots-Tag header, in 3 minutes, with the evidence included.