Written by Admin Alex · Fact-Checked by M.Ali · Info Verified September 2026
We review and update this article regularly as new information becomes available.
TL;DR: Google is testing a new “Dive deeper” button inside Google Discover that opens an AI-generated topic summary with links to related coverage, rather than sending the reader directly to the publisher’s story they tapped on. It’s starting with video content, and Google admits the AI summaries “can make mistakes,” but the bigger worry among SEO watchers is what it does to publisher click-throughs.
Every time Google adds a summary layer between a user and a publisher’s actual page, somebody in that publisher’s analytics dashboard has a bad afternoon. That’s the pattern with AI Overviews in regular search, and it’s the pattern SEO watchers now expect from Discover’s new “Dive deeper” button too.
What tapping the button actually shows you
Google product lead Robby Stein described the feature plainly when announcing the test.
“By tapping ‘Dive deeper,’ you’ll see a short topic overview with prominent links to related stories, community reactions and original reporting,” said Robby Stein.
In practice, that means a user browsing their Discover feed who wants more context on a story doesn’t land on the article itself first. They land on Google’s own AI-written summary of the topic, with links out to various sources bundled underneath it. Google is starting this test on video content specifically, with plans to try different designs over the coming weeks as it figures out how the feature should actually look and behave.
Why publishers should be nervous, according to at least one longtime SEO watcher
Search Engine Roundtable’s Barry Schwartz has been tracking this kind of change for years, and his read on “Dive deeper” is skeptical. If Google’s AI-generated overview already gives a reader enough context to feel satisfied, there’s simply less reason for that reader to click through to any individual publisher’s full article. The overview becomes the destination. The original reporting becomes a footnote link most people never open.
Google, to its credit, isn’t pretending the AI summaries are flawless. The company includes a disclaimer that the AI-generated content “can make mistakes,” the same hedge it uses across AI Overviews in search. That disclaimer protects Google legally and reputationally. It does nothing for a publisher watching referral traffic dip because fewer readers ever reach their actual page.
This is a test, but the direction is not new
Discover has historically been one of the more reliable traffic sources for news publishers, often sending more visits than organic search results for time-sensitive stories. Layering an AI summary step in front of that traffic source follows the exact same playbook Google has already run in search itself: introduce the AI layer as a small test on a narrow content type, expand it gradually, and let publishers adjust after the fact rather than being consulted beforehand.
Bottom Line: “Dive deeper” is early and limited to video for now, but publishers would be wise to assume it eventually expands the way every other AI summary feature Google has shipped in the past two years has expanded. If Discover traffic starts softening in the coming months, this button is the first place to look.
