noarchive
noarchive is a robots meta rule that asked search engines not to keep a cached copy of a page. Google no longer uses it.
noarchive is a rule you can put in a page's robots meta tag, or in its HTTP header, to ask search engines not to keep and show a cached copy of the page. It was written for the "Cached" links that search results used to carry.
You might still see it in your site's code or in an SEO plugin's settings. It's worth knowing what it does today, because it does less than its name suggests.
What it does now
Google has dropped cached links, and its robots meta tag documentation says it no longer uses noarchive to decide whether to show one. Other search engines still list the rule, and each one decides what it means, sometimes with a meaning for its own AI features. Check each one's documentation before you count on it.
At Google, that leaves noarchive with no job to do. It doesn't keep a page out of search results, and it doesn't stop the page from being used in an AI Overview.
It's easy to mix it up
The trouble with noarchive is that it sounds like a request not to store the page, which is close to what many site owners want from AI companies. So it gets added in the hope that it keeps AI models away.
For example, say Kestrel Roasters adds noarchive to its blog to stop AI bots from copying its articles. Nothing changes for the AI platforms. Their training bots and search bots still read the pages, because the rules they follow are in robots.txt.
If you want a specific result, use the rule that's documented for it:
- To keep a page out of Google's results, use
noindex. - To stop Google showing text from a page, or using it as direct input for AI Overviews and AI Mode, use
nosnippetormax-snippet. - To keep an AI company's bots off your site, block them by name in robots.txt.
In AnswerOrbit
The Audit tab in Website checks your robots.txt for each of the 42 AI bots we know. It doesn't read the meta tags on your pages, so a noarchive tag won't show up there.