The conversation has started. OpenAI launched a test phase for advertising inside ChatGPT last month, with Criteo as its first technology partner. Google is building capabilities for ads inside AI Overviews and AI Mode. Perplexity has retreated from the idea. And Anthropic — the company behind Claude — is refusing advertising entirely, on the grounds that the intimate nature of AI conversations makes commercial intrusion fundamentally different from a search result.
While the industry figures out what LLM advertising will look like, a more important question is going unasked: what happens to the brands that get cited organically, before any ad model exists — and after one does?
The monetisation debate misses the point
The Campaign Asia analysis of LLM advertising surfaces a tension that ad tech executives are only beginning to work through. One executive summarised it precisely: the monetisable unit in an LLM environment shifts from pages and impressions to intents, tasks and decision moments inside conversations. The primary performance event is no longer click-through rate. It becomes "recommended", "shortlisted", "chosen".
This is not a small shift. It is a structural redefinition of what advertising is — and what brand visibility means. In a world where the AI gives a direct answer, the value of being in that answer is orders of magnitude higher than being the third link on a results page. And the mechanism for getting there is completely different.
Paid inclusion in LLM responses — where it exists at all — will be labelled, contested, and limited to the platforms that choose to monetise this way. Organic citation, by contrast, is available on every platform, is unlabelled, and carries the full weight of the AI's recommendation behind it.
Anthropic's refusal is a signal, not an anomaly
Anthropic's decision to reject advertising is being framed in the industry press as a principled stand — and it is. But it is also a competitive signal that brands should read carefully. Claude is one of the most widely used AI assistants in professional and business contexts. There will never be a paid route to appearing in Claude's responses.
This means that for a significant and growing segment of AI-driven queries — particularly in B2B, professional services, and high-consideration consumer decisions — organic citability is the only available mechanism. Not a preferred mechanism. The only one.
The same logic applies, to varying degrees, across all LLM platforms. Even where advertising eventually becomes available, it will be labelled. Users will know. The unlabelled organic recommendation will always carry more weight — particularly as AI relationships become more personal and users become more sceptical of commercial intrusion in conversational contexts.
Clicks were never the point
One of the clearest insights from the current LLM advertising debate is that click-through rates in ChatGPT's early test phase trail Google's significantly. This is not surprising — it reflects something fundamental about how people use AI versus how they use search.
Search is transactional. Users expect to click. LLM conversations are consultative. Users expect answers. When an AI recommends a brand in a direct response to a question, the recommendation itself is the value — not the click that may or may not follow. Brand lift, recall, consideration, and eventual purchase all happen downstream of that moment of citation, not from a click event that measurement platforms can track in real time.
This is why the traditional advertising measurement framework — impressions, clicks, conversions — struggles to map onto LLM environments. And it is why brands that are waiting for a measurable paid channel before investing in AI visibility are misreading the landscape. The competitive window is not when LLM ads are measurable. It is now, before most brands have even started to think about citability.
What organic citation actually requires
Being cited organically by AI models is not a function of having a large advertising budget, a high domain authority, or a strong social media presence. It is a function of how well a brand's information is structured, distributed, and understood by AI systems.
AI models cite brands that are described consistently across authoritative sources, clearly associated with specific topics and use cases, and structured in ways that allow the model to extract and reproduce accurate information. A brand that ranks well in traditional search can still have a citation rate near zero if its content is ambiguous, inconsistent, or structured for human readers rather than machine processing.
The Citability Score framework measures exactly this readiness — across 56 parameters spanning identity and structured data, reputation and external network, content quality, semantic consistency, and technical performance. Most brands that have not been audited score between 30 and 45 out of 100. After a full Citation Rate intervention, scores typically reach 65 to 85.
The first-mover window is open — briefly
The Campaign Asia analysis notes that brands with strong AI visibility positions today are building a structural advantage that will compound over time. This is accurate, and the timeline is shorter than most marketing teams assume.
As LLM advertising becomes normalised, the contrast between paid and organic citation will become visible to users. Paid recommendations will be labelled. Organic recommendations will not. The brands that have invested in citability before the ad model matures will benefit from the full credibility of an unprompted AI recommendation — a form of brand endorsement that no paid format can replicate.
The brands that wait for a measurable paid channel will find themselves competing on a crowded and expensive paid layer, while the organic layer has already been claimed by earlier movers.
The measurement framework already exists
One of the recurring concerns in the LLM advertising debate is measurement. How do you track performance in a clickless environment? How do you attribute revenue to an AI recommendation that happened inside a private conversation?
These are real challenges for paid LLM advertising. They are not challenges for organic citation measurement — because the AI Visibility Index provides exactly this capability. The AVI tracks citation frequency, position, and sentiment across all major AI platforms — ChatGPT, Gemini, Claude, Perplexity, Copilot — over time, across the queries that matter most to a brand's category.
This is not a proxy metric. It is a direct measure of how often AI systems recommend your brand, in what context, and with what framing — before any click happens, before any conversion is logged, and regardless of whether the platform you are being cited on has an advertising model at all.
While the industry is still debating how to measure LLM ad performance, the measurement infrastructure for organic AI visibility already exists. The brands that are using it now are not waiting for the market to catch up. They are building the positions that the market will eventually try to buy its way into.
Source: Campaign Asia — LLM ads raise concerns: Experts discuss measurement and brand safety risks (March 2026).