Google AI publisher payments finally exist, and the first numbers show what a citation is worth when the platform is the only buyer: for many small and mid-sized sites, less than a tenth of 1% of their ad revenue. That figure is not a bug in an early pilot. It is the first public exchange rate between the click economy and the answer economy, and it was set with no advertiser in the room.
- Google's AI contribution pilot now pays about 100 publishers for content that shapes AI Overviews, AI Mode and Gemini answers, according to The Information.
- Reported payouts range from more than $1 million a year for one early entrant to under $1,000 over several months for small sites.
- The low rate reflects structure, not immaturity: the buyer sets the price, books it as a cost, and no advertiser is party to the transaction.
- Brands buying the ad slot beside the answer fund the platform, while the pages that build the answer are paid at the platform's discretion.
What the citation exchange rate is
The citation exchange rate is the value a platform assigns to content that shapes an AI answer, measured against what the same content earned when it drew a human click. In the open web, that rate was implicit: Google indexed the page, sent the visit, and the publisher monetized the session. In the answer economy, the visit often never comes, so the rate has to be paid explicitly or not at all.
AdSense economics vs AI contribution economics
AdSense, for all its flaws, ran on three things a market needs: an advertiser paying, a published split, and an auditable billing event. Google has long said publishers keep about 68% of AdSense revenue, and since its 2024 restructuring they receive 80% after the buy-side platform's fee. The click, and later the impression, was a unit both sides could count.
The AI contribution pilot inverts all three. No advertiser funds the payment directly, no share is published, and the billing event is a contribution score only Google can see. Participants get a monthly figure in Search Console with some history and no calculation breakdown, as Digiday reported on 14 September 2026. One model priced attention through an auction. The other prices it through a widget.
What The Verge reported on Google AI publisher payments, and what we verified
The Verge reported on 30 September 2026, citing The Information, that Google now pays roughly 100 publishers through a pilot that began less than a year ago. One early participant earns more than $1 million a year, a later entrant has made about $50,000 to $60,000, and several small sites collected under $1,000 over several months. Niche verticals such as anime and gaming reportedly earn more, and several larger publishers have declined to join.
A verification note: these figures come from different publishers over different time windows, as Search Engine Journal pointed out, so they describe a range, not a typical payout. The number that does generalize is the ratio The Information reported for many small and mid-sized participants: payments below 0.1% of their ad revenue. We anchor the analysis on that ratio, not on the $1 million headline.
Why a 0.1% payout is the market working as designed
The case that a pilot payout is a floor, not a verdict
The strongest defense of the pilot is real. Google spent two decades refusing to pay for search inclusion, arguing that traffic was the payment. It now pays per contribution, the usage-based logic publishers have demanded for years. One participating executive told Digiday they see the pilot as an early test of a marketplace for inference data, and opening prices in a new market are rarely the long-run price.
Why a single-buyer payout stays a cost line
The problem is who sets the number and where it sits on the ledger. Google sells ads around AI Overviews and AI Mode, and the contribution payment is not a share of that revenue. It is a discretionary expense, and expenses get minimized. Participants told The Information their payments swing month to month without explanation, which is exactly what a cost line managed by the payer looks like.
Now look at the demand side. OpenAI launched ChatGPT ads in February 2026 at a reported $60 CPM, and market rates have since settled roughly between $25 and $60. Advertisers pay premium prices for the slot beside the answer, while the sources that build it receive a sliver of their old economics. People Inc. CEO Neil Vogel described the fear at a Fortune event in September 2025: analog dollars became digital dimes, and dimes are becoming AI pennies. The pilot just put a decimal point on it.
Seen this way, the large publishers holding out are not being difficult. They are refusing to let a single buyer's opening number become the industry's reference price for AI content.
What this means for brands and for publishers
For CMOs, media buyers and agencies: you are buying the slot, not the source
Every dollar spent beside an AI answer buys platform inventory, not the pages that decide whether your brand appears in the answer at all. When the platform also pays those pages on its own terms, their incentives drift toward the platform's surfaces, not toward you. Map which publisher pages ground answers in your category, buy presence on them directly, and ask every AI ad seller for source-level contribution data, not just impressions.
For publishers: a Search Console widget is not a rate card
Take the pilot money if it is offered, but treat it as market data, not a business model. Log contribution payouts against the ad revenue the same pages generated before AI Overviews, and you will know your own exchange rate. Then sell the asset Google is underpricing: being the source AI engines rely on, sold to the brands that need to appear in those answers, at a price set by more than one buyer.
Three signals to watch over the next six months
First, regulators: the UK competition authority ruled in June that Google must let publishers opt out of AI search features, and the European Commission is investigating, so a forced disclosure of the contribution formula would reset the debate. Second, the holdouts: if a major publisher group joins on undisclosed terms, the reference price is set. Third, the buy side: watch whether any AI ad seller gives advertisers visibility into the sources behind the answer their ad sits beside. That is the moment demand starts pricing supply.
Conclusion
Hold on to this: Google AI publisher payments did not reveal what a citation is worth, they revealed what a citation is worth when its only buyer sets the price and no advertiser is in the transaction. That gap is where Generative Engine Advertising belongs: Smalk AI places native ads for AI agents on the publisher pages AI engines rely on, so brands pay to be present where answers are built and publishers are paid by demand rather than by a discretionary widget. What to watch next: whether the first major publisher to sign sets the citation exchange rate, or whether a brand-funded market sets it first.
