Google Drops AI Overview Labels on Desktop for Some Results

Google is showing some AI Overviews (the AI-generated answer blocks at the top of Google Search results) without their "AI Overview" label on desktop, while the same results carry the label on mobile. SEO consultant Glenn Gabe spotted the discrepancy on the morning of October 20, 2026, and posted his finding on X.

Gabe wrote: "No AIO label? -> Very interesting to see AIOs for shows or movies and Google not provide the AIO label. Noticed that this morning. I wonder if we will see more AIOs without labeling. Stay tuned."

The unlabeled results are not the standard link-based answer blocks most search users have grown accustomed to. They appear in knowledge-panel-style AI Overviews — the boxes Google serves for entities such as TV shows and movies, which sit at the top of the results page and summarize information the way a knowledge panel would.

What exactly is missing?

The core facts of the sighting break down cleanly:

• Desktop AI Overviews for certain shows and movies display no "AI Overview" label at all.
• The identical results on mobile retain the "AI Overview" label at the top of the box.
• The overwhelming majority of AI Overviews across Google Search still carry the label as normal.
• Google has not yet commented; Google was contacted for a response.

The desktop-versus-mobile split is the detail that matters most here. A deliberate interface change would ordinarily roll out consistently across surfaces, or at least appear somewhere in Google's steady stream of feature announcements. A rendering inconsistency between desktop and mobile for the same result strongly suggests an engineering slip rather than a policy shift.

Is this a bug or a deliberate change?

The evidence points toward a bug, and the reporting around it says so explicitly. Google has previous form: AI Overviews have lost their labels before, and on those occasions the missing labels turned out to be a bug that Google subsequently fixed. The current incident matches that pattern — sporadic, entity-specific, and inconsistent across devices.

That said, nobody outside Google can confirm intent yet. Google has not responded to the outreach, and until it does, the bug theory remains informed speculation rather than a confirmed statement from the search engine. The confirmed facts are limited to what appears on the results page: some desktop AIOs (the common shorthand for AI Overviews) are unlabeled, and their mobile counterparts are not.

Why unlabeled AI Overviews matter

Labels are not cosmetic. The "AI Overview" tag tells users that the content they are reading was generated by a language model summarizing web sources, not curated by a human editor or drawn directly from Google's knowledge graph. Strip the label, and a casual user has no reliable way to distinguish an AI summary from a traditional knowledge panel — a meaningful distinction at a moment when publishers and regulators are pressing AI companies on attribution and transparency.

What does this change for site owners in practice? Very little, if it is a bug. Traffic measurement, AI Overview tracking tools, and click-through patterns should not shift because a label failed to render. But if unlabeled AI Overviews ever became deliberate and widespread, third-party tracking would grow harder, because many tools identify AI Overviews partly by their label markup. Quiet removal would also raise fresh transparency questions just as AI answers occupy more of the results page.

A possible explanation for Semrush's tracking dip

The timing of the label gap lines up with a separate data point. Semrush, whose sensor tools track how often AI Overviews appear in Google Search, is currently reporting a drop in AI Overview prevalence.

Search Engine Roundtable editor Barry Schwartz connected the two observations on X the same day, writing: "@semrush reporting a drop in how often AI Overviews show up, this has to be a bug in Semrush tracking - I see AIOs everywhere."

The logic is straightforward. If some AI Overviews are rendering without labels or with broken markup, automated tracking systems that detect AIOs by their labeled structure will undercount them. Schwartz's own manual checks — he reports seeing AI Overviews everywhere — contradict the Semrush decline, which supports the tracking-artifact explanation over an actual reduction in AI Overview coverage.

Here again the line between confirmed and inferred matters. What is confirmed: Semrush's data shows a drop. What is inference, explicitly flagged as such by Schwartz: that the drop stems from a detection bug tied to rendering changes rather than from Google actually serving fewer AI Overviews.

What to watch next

Monitor three things. First, whether Google responds and confirms the missing labels as a bug, as it has in past incidents. Second, whether the unlabeled results spread beyond desktop show-and-movie panels to general informational queries — that would be the signal of a deliberate test rather than a glitch. Third, whether Semrush's AI Overview tracking recovers once labels render correctly, which would confirm the measurement artifact. Gabe himself put it best: stay tuned.