How Often to Update Content for AI Search (2026 Answer)

Update a page when a claim on it has expired, not when a calendar says so. The cadence tables circulating in this category — product pages monthly, blog posts quarterly, everything annually — are not traceable to a study, and the study they are usually justified by says something narrower than the tables do.

Here is the cadence, before the evidence:

This survives either outcome of the freshness question. If freshness is causal, you updated the pages that had reason to change. If the 25.7% turns out to be an artefact, you still corrected content that had become wrong.

How often should I update content for AI search?

There is no published cadence to follow. The largest study reports that AI-cited pages average 1,064 days old against 1,432 days for organic results — 25.7% "fresher" — and in the same post says cited pages are still 2.9 years old on average and that Google is the surface least influenced by freshness. Update on claim expiry, not on a timer.

Marks used below: △▲ observational study, no control group, published by a company selling a related product. ○▲ preprint with a statistical result, vendor-affiliated authors. ◐ our own first-party measurement. ✗ quoted widely, no method we could locate.

The number everyone quotes, traced

Ahrefs, by Ryan Law and Xibeijia Guan, published 28 July 2025. Method as stated: 16,975,000 cited URLs collected via their Brand Radar product from ChatGPT, Perplexity, Gemini, Copilot and AI Overviews, plus organic Google SERPs for comparison, with both publication date and last-updated date recorded. Grade △▲ — observational, no control, and Ahrefs sells the tool the data came from.

The average age of URLs cited by AI assistants is 1064 days, compared to 1432 days for URLs in organic SERPs — 25.7% "fresher."

That is a real comparison with a stated denominator, and it is the strongest thing in this area. Note precisely what it is: a difference between two average ages. It is not a measurement of what happens to a page when you update it.

If you have seen it quoted as "AI search relies 26% more on recent information" — that is a paraphrase, and it changes the claim. "The pages it cites are on average 25.7% younger" and "it depends 26% more on recency" are different sentences. The second implies a weighting inside a system nobody outside Google or OpenAI can see.

The same post contains the correction

This is why it is worth reading rather than quoting. The authors state, in their own words:

• "The average age of cited pages is still 2.9 years." Like traditional search, AI assistants still prefer citing long-lived content.
• "Google is the least influenced by content freshness" — and Google is still where the majority of searches happen.
• "Content freshness is one among many factors. Low-quality, irrelevant content that's updated every day will not have a magic positive effect."
• They explicitly warn against updating publish dates without changing the content, citing Google's John Mueller.

So the honest one-sentence summary of the largest study here is: pages cited by AI assistants skew younger than organic results, by about four months on a base of nearly three years.

A second study: refreshing is most of the freshness

Seer Interactive, published 2026, grade △▲ (a marketing agency selling work in this area). Method as stated: 7,683 pages carrying 47,097 citations, March-June 2026, across ChatGPT, Gemini and Perplexity, in four industries. Two-thirds of pages had readable last-modified dates from schema, sitemaps and headers.

The last three rows are the most useful lines published on this topic. The gap between 72% and 42% is the entire practical question: if you look at publish dates most cited content looks old, and if you look at update dates most of it looks new — and these are the same pages.

Roughly a quarter to a third of everything that looks fresh in the citation set is old content that was updated, not new content that was written. Refreshing is not a lesser substitute for publishing here. It is most of the freshness.

Note the engine spread too — 78% to 65% — the same shape as everything else in AI search. The surfaces are not one thing.

The third study measured decay rather than stock

Scrunch AI with Stacker, published March 2026. Method as reported: over 3 million citation events, eight industries, six AI platforms, tracked across 26 weeks, analysed with cohort-based survival analysis and 200 bootstrap simulations. Headline: the median citation half-life for non-distributed, single-domain content is 4.5 weeks. Grade △▲ — Scrunch sells AI visibility software, Stacker sells content distribution, and the finding is flattering to precisely what Stacker sells.

Survival analysis is the right instrument for this question and, as far as we can tell, nobody else in this area is using it. Average age describes the stock of cited pages at a moment. A half-life describes the rate at which an individual page falls out, which is what a refresh cadence is trying to counter.

We have not opened the original. We read it through secondary write-ups and are reporting the method as they describe it. Until we have read the primary, we would not put weight on 4.5 weeks as a value — only on the observation that somebody has finally measured decay.

The 13-week number, and why we grade sources

You will meet the claim that 50% of AI citations come from content under 13 weeks old. It is the single most repeated figure on this topic. Tracing it, we found it credited to four different origins by different republishers — Seer Interactive, Ahrefs, Scrunch/Stacker, and an analysis by Lily Ray presented at Tech SEO Connect 2026.

We read the Ahrefs post directly and it contains no 13-week figure. It reports average ages of 1064 and 1432 days. So at least one of those four attributions is wrong, and a reader has no way of telling which without repeating the work.

This is not us calling the number false. It may be perfectly good. The problem is structural: a figure whose provenance changes depending on who repeats it cannot be checked, which means it cannot support a decision you will be held to.

Still untraced after looking:

The confound none of them control for

AI citations and organic results are not drawn from the same population of page types. Comparing their average ages compares two mixtures, and a difference in the mixtures produces a difference in average age with no freshness preference existing anywhere in the system.

Our own citation data makes that concrete rather than theoretical ◐ — one brand, one category, July 2026:

Now consider how age behaves per genre, before anyone optimises anything:

• A "best tools 2026" listicle carries the year in its title and gets re-dated annually as routine. It is fresh by construction.
• A YouTube page has a publish date and effectively never gets updated — but the ones surfacing for a live query tend to be recent uploads.
• A vendor product page may be edited constantly and carry no visible date at all.
• A community thread is dated by its newest reply.

If AI citations over-select listicles and video relative to organic results, the citation set will look younger even if freshness is not a factor at all. Sorting by age would recover the genre mix, not a preference.

This confound is unproven in both directions. Settling it means holding genre constant — listicles against listicles, product pages against product pages. Neither published study reports doing that. What the table establishes is the precondition: the genre mixes differ sharply by surface, which is all the confound needs in order to bite. The engine split behind that table is in Brand Not Showing in AI Search?, and the YouTube row is unpacked in Does YouTube Help AI Search Visibility?.

Why a naive before-and-after won't settle it either

Suppose you skip the studies and just test it: update thirty pages, re-measure, see if citations rise.

That design produces a number, and the number is close to meaningless, because the citation set churns on its own. Published measurement puts the overlap between consecutive days' cited sources at a Jaccard index of roughly 0.336-0.423 — around 65% of cited sources change overnight with nobody touching anything (Schulte et al. 2026, arXiv 2604.07585, grade ○▲). At a single sample, the 95% interval on a brand-detection rate is about ±72 percentage points.

Two numbers drawn once from each end of a distribution that wide will contain almost any story you want to tell.

The test that would settle it

1. Matched pairs, not a single group. Pair pages by genre, topic and current appearance rate. Update one of each pair; leave the other alone. The untouched half is the whole experiment — it absorbs the overnight churn.
2. Repeated sampling at both ends. A rate per question per engine, never a single observation, never pooled across engines.
3. Substantive updates only, and log what you changed. Seer's 72%-versus-42% gap means "date changed" and "content changed" are separable in the data. If your treatment is a date bump, you have tested date bumps.
4. Hold genre constant. Do not update your listicles and compare against untouched product pages. That reproduces the confound inside your own test.
5. Fix the observation window in advance. Nobody has published how long an intervention takes to reach these surfaces. We do not know either.
6. Publish the null result if you get one. This field currently has a publication filter on it: the studies that exist were run by companies selling adjacent products, and a null is worth nothing to them commercially.

Nobody has run this test, including us. Everything above is observation on citation sets, not intervention on pages.

What we would not do

Bulk date-bumping. Scheduled republishing with no substantive edit. Rewriting an evergreen explainer on a quarterly timer because a table on a vendor blog said to. The one thing multiple sources here agree on — including the authors of the largest study — is that changing the date without changing the page is the version that does not work.

The part of this we did build

There is a smaller problem hiding inside "update the page": knowing whether the edit is actually live on the page an engine reads. Staging, caching, a component that renders the fix only after JavaScript runs, an edit that never got deployed — all of these look identical to a finished task in whatever list you keep.

That is the loop peekr closes. The page check returns concrete items across AI crawler access, AI legibility and on-page SEO. You can mark an item as fixed, and marking it does not remove it. It leaves the list when a later check re-reads the page and no longer finds it. If you marked something and the next check still sees it, the tool says so in those words: either the change is not live on the page we read, or it is somewhere this check does not look.

The tool is also explicit about the limit this whole post is about: everything on that list being gone does not mean AI mentions you more. Only asking the engines tells you that, and that is a separate measurement with a separate number — see How to Track Brand Mentions in AI Search.

[Run the page check on one URL](/en/onboarding/preview) One site, no account.

Method

External figures are attributed inline with publisher, date, method as stated by the publisher, and a conflict-of-interest mark. We read the Ahrefs and Seer Interactive write-ups directly; we did not receive or inspect either underlying dataset, and neither is public as far as we can tell. Figures marked ✗ are ones we searched for and could not trace to any stated method — we have listed rather than used them.

Figures marked ◐ come from our own stored measurements on one brand in the AI video-editing category, July 2026: Google AI Overviews retrieved through a SERP API across 13 queries, and engine responses through the OpenAI and Google APIs with retrieval enabled, producing 154 citations in the head-term query set. Raw responses retained. We do not crawl third-party sites.

If you have run a matched-pair freshness test — in either direction — I would like to read it, and will link it here.