For months, appearing in an AI Overview or AI Mode response felt like a binary outcome - you were either cited or you weren't. That is changing. Google is now officially rolling out preferred source labels within AI Overviews and AI Mode, introducing a new carousel format in both surfaces, and expanding the highly cited label to a wider set of search results. These are not cosmetic tweaks. They represent a visible hierarchy of source authority that users will see and respond to.
The practical implication is significant: being cited in AI search is no longer the finish line. Where you appear, how you are labelled, and whether you carry a trust marker will all influence whether a user actually clicks through. Brand visibility strategy now has to account for these layers.
What Preferred Source Labels Actually Signal
A preferred source label, as Google is rolling it out, tells the user that this source has been identified as particularly relevant or authoritative for the query at hand. It is a direct, visible editorial signal from Google appearing inside the AI-generated response itself. Users do not need to interpret relevance signals the way they once did with position or domain authority - the interface does it for them.
This matters enormously for click-through behaviour. In traditional search, users developed their own heuristics for evaluating results. In AI search, Google is increasingly pre-interpreting authority for them. A brand that earns preferred source status on a category of queries gains a visibility advantage that is explicit rather than implied.
The question for brands is what drives that label. While Google has not published a definitive formula, the broader context points to the same factors that determine citation frequency in the first place - content depth, topical consistency, structured information, and signals of real-world expertise. There is no shortcut to preferred source status. It is the output of sustained content credibility, not a metadata trick.
The Highly Cited Label Is Now a Mainstream Signal
Alongside the preferred source work, Google is expanding the highly cited label to more search results - and this expansion is not limited to AI Mode or AI Overviews. It appears across broader search surfaces. This is worth sitting with for a moment. A signal that emerged in the context of AI-generated responses is now being applied more widely, which tells you something about the direction of travel for how Google evaluates and surfaces content authority.
For UK brands managing AI visibility, this creates a concrete goal to work towards. Being highly cited does not just improve your standing inside AI responses - it now carries a visible label in standard results too. That creates a compounding effect. The same content quality and authority-building work supports your presence across multiple search surfaces simultaneously.
Think about what earns citation frequency in AI systems: clear answers to specific questions, structured content that is easy for models to parse, consistent brand voice across sources, and third-party references that validate your claims. These are not abstract principles - they translate directly into the kind of content architecture that brands need to be building now.
The New Carousel Format Changes the Visibility Equation
The new carousel being added to AI Mode and AI Overviews introduces a different visual structure for how sources are presented. Carousel formats in search have historically concentrated click attention on a small number of visible items. Users scroll through them selectively, and the first few positions receive disproportionate engagement. There is no reason to expect AI Mode carousels to behave differently.
This changes the competitive picture inside AI search results. When citations were presented as a flat list of links, there was at least surface parity between sources that appeared together. A carousel format introduces ordering, visual prominence, and the possibility of brand imagery or structured metadata playing a role in how entries appear. Brands that have invested in clean structured data, strong entity definitions, and consistent visual assets will be better positioned to benefit.
For sectors where comparison is natural - products, services, local businesses, financial options - carousel placement in AI Mode could become one of the most commercially significant visibility positions in search. Getting into the carousel is one challenge. Appearing prominently within it is another. Both are worth treating as distinct strategic objectives.
What This Means for a GEO Strategy in Practice
The combined effect of these changes - preferred source labels, expanded highly cited tags, and carousel formats - is that AI search is developing a more explicit quality tier. Brands that have been treating AI visibility as a binary yes/no question need to update their thinking. The question is now: what tier are you in, and what does it cost you to be in the wrong one?
Practically, this means auditing which queries your brand currently appears in AI responses for, and then assessing whether you carry any trust label or preferred status on those queries. If you are being cited without a label while competitors carry one, you are losing ground even when you show up. That is a new kind of competitive disadvantage that did not exist six months ago.
The content work required is not fundamentally different from what good GEO strategy has always demanded - authoritative, well-structured, genuinely useful content that earns third-party references. But the stakes for getting it right have risen. These labels are user-facing. They will influence trust and click decisions directly, and they will do so in a search environment where attention is already scarcer than it was.
The Strategic Shift: From Appearing to Being Trusted
There is a clean way to frame what Google is doing here. It is moving AI search from a model where brands compete to be included in responses, to one where brands compete to be visibly endorsed within them. That is a meaningful shift in what the work involves.
Preferred source status and highly cited labels are public-facing trust endorsements served inside the search interface itself. Earning them requires the same foundations that have always determined genuine content authority - expertise, consistency, real-world credibility, and a clear topical identity. But now those foundations produce a visible competitive differentiator rather than an invisible ranking factor.
Brands that have been patient with their AI visibility investment are about to see the distinction between those who built real authority and those who optimised for surface-level citation. Preferred source labels will make that distinction legible to users. That is both an opportunity and a risk, depending on where your content programme stands today.