Does AI Search Prefer Fresh Content? Read the 25.7% Figure Carefully

AI-cited content was younger on average than organic-search content in a large Ahrefs analysis. That does not mean updating a page creates a predictable increase in citation probability.

Freshness matters when the information changes. It should not become a reason to rewrite accurate content without improving it.

What did Ahrefs measure?

The July 2025 analysis examined approximately 17 million cited URLs across several search and AI experiences. It used crawler first-seen dates as a proxy for publication and database dates for updates.

Its headline comparison reported an average age of 1,064 days for URLs cited across ChatGPT, Perplexity, Copilot and Gemini, versus 1,432 days for organic results—a 25.7% difference.

The article’s detailed table uses a slightly different organic-publication average of 1,416 days. That inconsistency is a reason to attribute the headline result precisely rather than present it as an exact universal benchmark.

Younger does not mean newly published

An average age of 1,064 days is about 2.9 years. Even the AI-cited content in the headline comparison was not predominantly yesterday’s news.

The study also reported variation between platforms. Google AI Overviews showed a stronger association with older content than several assistants.

A single rule such as “update every page monthly for AI” therefore ignores both topic differences and platform differences.

What cannot be inferred?

This was an observational comparison. It did not randomly update otherwise equivalent pages and measure the resulting change in citations.

Younger pages might cover different topics, contain newer research or belong to different publishers. The study does not isolate freshness from all those other characteristics.

Nor does a 25.7% difference in average age mean a 25.7% gain in visibility after an update. Those are different quantities.

What does meaningful maintenance look like?

Review pages when products, prices, integrations, regulations, source data or recommended methods change. Correct broken examples and misleading claims. Add new evidence when it improves the answer.

For evergreen explanations, the review may confirm that the content remains accurate. Record that internally without pretending the page has been substantially rewritten.

Keep publication history honest. Do not reset dates to manufacture recency or imply new research where none exists.

Prioritise by risk and commercial importance

A pricing page or integration guide may need attention as soon as a change occurs. A conceptual explanation may remain useful for much longer.

Maintain a content register with an owner, topic, source dependencies, review trigger and last meaningful change. This makes maintenance accountable without turning it into a word-changing exercise.

Google recommends helpful, reliable content for its AI features; it does not prescribe a universal rewriting interval.