Google adds Top Stories to AI Overviews, reshaping news visibility

On July 17, 2026, Google confirmed that Top Stories were rolling out inside AI Overviews. With Top Stories appearing inside the overview itself, one search page can hold the summary, the cited sources, and the news carousel before a reader ever reaches a blue link. Visibility is no longer won only by ranking first. It is now split across the answer, the source set, and the fast-moving news module.

The new search layer

The feature was later fully live for U.S. mobile users only. Google treated wider expansion as intent rather than a commitment, which matters because publishers cannot plan around a feature that is still being staged by device and market.

Digital Applied tied the Top Stories layer to Google’s Preferred Sources and freshness signals rather than to a new ranking factor. This is not a fresh scoring system publishers can reverse-engineer as a standalone trick. It is Google using the signals it already trusts, then showing news inside the answer layer where the user’s attention is most compressed.

Why the click path is getting tighter

Top Stories inside AI Overviews change the economics of a search results page because they reduce the number of independent choices a reader sees before getting an answer. Once the overview appears, the user is no longer choosing between a list of headlines in the same way they did in classic search. They are seeing a synthesized response, supporting citations, and news links all in one frame.

Pew Research Center’s July 22, 2025 finding lined up with that pressure: Google users were less likely to click on links when an AI summary appeared in the results. The Pew survey covered 900 U.S. adults in March 2025, and users clicked on links within the AI summary itself only 1% of the time. Google can surface your work inside the answer layer and still keep most of the traffic for itself.

Preferred Sources is now part of the same fight

Google extended Preferred Sources into AI Overviews and AI Mode on May 27, 2026. Google had already logged more than 345,000 unique sources selected for Preferred Sources, and preferred sources were being clicked roughly twice as often, according to Google.

If your newsroom or brand becomes a preferred source, you do not just gain exposure in a hidden settings panel. You raise the odds of being pulled into the answer layer and the news layer at the same time. If you are absent, the system still works, but it works with someone else’s reporting.

What this means for newsrooms and brands

By July 30, 2026, Google was adding Top Stories links to many news-related AI summaries. Strong reporting, quick indexing, and topical authority matter more because the window to be included is narrower. A story that arrives late, lacks clean structure, or cannot be crawled and interpreted fast enough will lose the slot before it ever gets a chance to earn a click.

The Reuters Institute-linked expectation in 2026 was that search traffic would fall by 43% in the next three years. Search Engine Land in January 2026 put the expected drop in search referrals at 43% by 2029. Digital Content Next had already warned in August 2025 that Google’s push to AI was hurting publisher traffic.

How to compete inside AI Overviews

Competing here is operationally demanding. Newsrooms and brands need to publish for fast inclusion, not just searchability. That means tighter headlines, clearer article structure, faster updates, and the kind of source authority Google is already rewarding through freshness and Preferred Sources.

A practical approach now looks like this:

• Build stories that can be crawled and summarized cleanly, with obvious entities, dates, and outcomes in the opening paragraphs.

• Treat freshness as a publishing asset, not a bonus. Top Stories inside AI Overviews are built to favor timely coverage and are tied directly to freshness signals.

• Maintain a clean Preferred Sources strategy. Google has already turned that setting into a measurable discovery channel, with more than 345,000 unique sources selected and preferred sources clicked roughly twice as often, according to Google.

• Write for the answer layer and the click. If a reader sees your story inside an AI Overview, the headline, subhead, and opening must earn the next action immediately.