Google AI Mode Now Cites Its Own Pages More Than Almost Any Other Domain

Google.com has quietly climbed to become the second most-cited domain in its own AI Mode search product — raising fresh questions about self-referential AI responses and what that means for brands trying to win visibility in the new search landscape.

According to a new report from Profound, an AI-visibility platform that tracks brand presence in AI search, citation share for google.com rose 8.4 times between April 15 and June 30, 2026. Nearly all of that increase flowed into two Google-hosted surfaces: Business Profile cards and Product Knowledge Panels.

---

What the Data Actually Reveals

Citation share measures how frequently a domain appears as a source in AI Mode responses. When users ask AI Mode a question, the system generates an answer and attaches citations pointing to the sources it drew from. Google.com now holds the second position on that list — behind only a handful of independent publishers.

Profound's analysis identified two specific subpaths driving the surge. The first delivers Business Profile cards for local queries. The second provides Product Knowledge Panels for product-related searches. Outside of these two surfaces, the data shows little growth in citations pointing to Google's organic results pages.

The concentration of citations is not evenly spread across all query types. For local searches, five sectors account for the most notable change:

• Hospitality and travel
• Home services
• Restaurants and dining
• Real estate
• Healthcare

For product searches, the pattern shifts to queries involving comparisons, compatibility checks, and specifications — the kind of research a shopper conducts before making a purchase decision.

"These cards are a lever built from data you control," the Profound report notes, pointing to Business Profile information for local results and product feeds for shopping queries. That framing offers a more actionable read on the trend than a straightforward narrative of Google crowding out independent sites.

Why This Matters Beyond the Numbers

Understanding the mechanics here is important. AI Mode does not simply reproduce web results — it synthesises information and selects citations to substantiate its generated answers. When Google's own hosted surfaces consistently appear among those citations, the effect is structural, not incidental. Brands and publishers are not competing on an open playing field; they are competing within an environment where one participant also controls the infrastructure.

To understand how AI systems like this make decisions about sourcing and citation, it helps to have a working knowledge of how artificial intelligence actually works and what drives its outputs — something that is becoming increasingly relevant for anyone managing digital visibility.

---

A Pattern With a Documented History

This is not the first time researchers have flagged Google's tendency to cite its own properties in AI-generated responses. The current data fits into a trajectory that SEJ has tracked for more than a year.

In March 2026, SE Ranking data reported by Search Engine Journal showed that Google accounted for 17% of 1.3 million citations analysed in AI Mode. That figure climbed to approximately 20% when YouTube was included in the count. SE Ranking and Profound used similar citation-share metrics but analysed different datasets and timeframes. Both arrived at the same broad conclusion: Google properties appear in AI Mode citations with striking regularity.

The pattern goes back further still. In May 2025, SEJ reported that 43% of AI Overview responses referenced Google's own results — an earlier form of the same self-citation behaviour now surfacing in AI Mode.

What the current Profound data adds is specificity. Rather than pointing broadly at google.com, the report isolates which surfaces are driving growth and which query categories are most affected. That granularity matters for marketers trying to understand where their visibility is going and why.

The Regulatory Context

The legal and policy environment surrounding this behaviour is unsettled and developing rapidly. The European Union's Digital Markets Act and ongoing antitrust proceedings in the United States both have potential bearing on how Google can use its own surfaces as citation destinations inside a product it controls. Regulators in both jurisdictions have already demonstrated willingness to scrutinise Google's market conduct in search. Self-citation within a generative AI product — particularly one that may soon facilitate commerce — is precisely the kind of behaviour likely to attract closer examination.

For now, no formal findings have been made specific to AI Mode citation patterns. That may not remain the case. Brands and publishers monitoring this space should follow regulatory developments around AI in search as closely as they follow the technical shifts.

---

Product Panels, Commerce, and What Comes Next

The Universal Commerce Protocol Connection

Profound raises an additional layer to the product-panel story. The firm sees the rise in Product Knowledge Panel citations as an early signal of Google's Universal Commerce Protocol — an open standard Google has described as enabling direct purchases across AI Mode and Gemini, with retailers serving as the Merchant of Record.

That connection is Profound's interpretation rather than a confirmed finding from Google. The report stops short of stating that the panels appearing in its dataset are UCP-enabled. Still, the framing points toward a future in which AI Mode does not simply surface information about products but facilitates the transaction itself — with Google's infrastructure sitting at the centre of that exchange.

If Google functions as a commerce layer inside an AI-generated response, it will face scrutiny from regulators already examining the company's market position in search. The implications extend well beyond citation metrics — they reach into how revenue flows, who owns the customer relationship, and where liability sits when a transaction occurs inside an AI interface.

What Brands Should Do Right Now

Citation share indicates what AI Mode links to — not where traffic ultimately goes. An increase in Google.com citations does not automatically mean a visit to an independent website was replaced. What it does mean is that for local and product queries, AI Mode may cite a Google-hosted card displaying a business's hours, photos, reviews, or product specifications instead of linking directly to that business's own website.

For publishers, the dynamic is familiar. Self-citation patterns redirect the citation to Google's surface rather than an outside site. For brands, the picture is more nuanced because the card typically reflects data the brand itself supplied.

Optimising a Business Profile or maintaining an accurate product feed is more controllable than attempting to rank an entire website. However, that optimisation only secures presence inside the cited card. It does not guarantee clicks or downstream sales.

Businesses that already use Google's core tools to grow their online presence are better positioned to act on this shift — the same platforms used for search visibility now directly influence what AI Mode cites.

Three practical steps worth taking immediately:

1. Audit your Google Business Profile and product feed data. Inaccurate hours, missing photos, or outdated specifications will appear in AI Mode citations exactly as submitted. The quality of your structured data is no longer just an SEO consideration — it is your content inside the AI response.
2. Monitor citation share for your brand and your competitors using AI-visibility tools. Citation presence is becoming a measurable benchmark alongside traditional ranking metrics.
3. Watch the Universal Commerce Protocol closely. If Google expands UCP integration into product panels, your feed data could influence not just citations but actual purchase pathways inside AI Mode.

The Wider Shift in Search Visibility

The broader takeaway is that the rules of search visibility are shifting beneath the surface. Brands that treat structured data and owned profiles as afterthoughts are increasingly handing Google the content it needs to answer queries — without sending users anywhere else.

The organisations most exposed to this shift are those that have relied on organic search traffic without investing in the structured, machine-readable data layers that AI systems prefer. As businesses across sectors adopt AI-driven tools and platforms, the gap between those with clean, optimised data infrastructures and those without will only widen — and AI Mode citation patterns are one of the earliest visible symptoms of that divide.