What OpenAI Reported at Cannes Lions
Speaking alongside CNBC's Julia Boorstin, OpenAI's chief revenue officer Denise Dresser said relevance has visibly improved since the company began testing ads in ChatGPT in February 2026, framing the falling dismissal rate as proof that the format is maturing rather than irritating users (Adweek). On the same stage, David Dugan described ChatGPT users to advertisers as "super intentional," noting that close to one in five ChatGPT conversations already carries commercial intent and that OpenAI sees ads playing a role in the research and comparison stage of a purchase, not only the final click (Marketing Dive).
That framing matters for anyone running AI visibility strategy. OpenAI is not pitching ChatGPT ads as a replacement for search advertising; it is pitching them as a new placement inside the moment a buyer is actively comparing options, the same moment brand visibility work has always targeted. The difference is that this moment now has a price tag attached to it on at least one major AI platform.
How the ChatGPT Ad Auction Works
ChatGPT ads run on what OpenAI calls a relevance-weighted, second-price auction. Each advertiser sets a maximum CPC or CPM bid, but the system combines that bid with a relevance score before ranking competitors, and the winner pays one cent more than the next-highest effective bid rather than its own maximum (PPC Land). OpenAI recommends starting CPC bids of $3 to $5, and the original CPM model priced impressions at $60 per thousand.
Targeting does not run on keywords. Advertisers set "context hints," natural-language descriptions of the conversations where their product is genuinely useful, and OpenAI's matching system weighs those hints against the live conversation, the user's prior chats, and past ad interactions. A tightly accurate context hint can beat a larger budget because relevance is a multiplier on the bid, not a tiebreaker.
This is the detail most coverage of the Cannes announcement missed: relevance, in OpenAI's own auction language, is functionally the same concept that decides whether your brand gets cited for free in an organic ChatGPT answer. The mechanics are different (an auction model versus a retrieval and generation pipeline) but both systems are scoring how well-described and how relevant your brand is to the conversation at hand. A brand with a thin or inconsistent entity record pays a relevance tax twice: once in lost organic citations, and now again in higher effective CPCs.
Ads Sit Beside the Answer, Not Inside It
Ads appear as a separate, labeled unit beneath ChatGPT's organic response. The model generates its answer first, using the same retrieval and reasoning process it would use without an ad system attached, and a distinct ad-serving layer then decides whether a sponsored placement is eligible to appear alongside that answer. Buying an ad does not change what the model says, who it names, or which sources it cites in the response itself.
That separation is the most important strategic fact in this announcement. A brand could in theory run a high-spend ChatGPT ad campaign and still never be the brand ChatGPT recommends organically when a user asks an open-ended "what's the best option" question with no commercial box attached. Ads buy a placement next to certain high-intent queries. They do not buy your way into the model's own answer, and they do not exist at all on Gemini, Claude, or Perplexity, where the only path to appearing in a response is still earning a citation or a named mention.
Organic Citation vs. Paid ChatGPT Ads
| Dimension | Organic citation / mention | ChatGPT ads |
|---|---|---|
| How you win it | Entity clarity, third-party authority, content structure | Relevance-weighted bid (CPC $3 to $5, or CPM ~$60) |
| Where it appears | Inside the model's own answer text | Labeled box beneath the answer |
| Platforms it works on | ChatGPT, Gemini, Perplexity, Claude, Copilot, AI Overviews | ChatGPT only, in 8 countries as of June 2026 |
| Time to see results | Weeks to months (entity and authority signals compound) | Live within days of campaign launch |
| Persistence | Compounds; does not require ongoing spend | Stops the moment spend stops |
| Trust perception | Read as the model's own judgment | Read as a paid placement, labeled as such |
The two channels are not substitutes. A brand with strong entity signals and weak ad spend still gets named in ChatGPT's free responses. A brand with heavy ad spend and a thin entity record still gets skipped in the organic answer, and now pays more per click for the privilege of the sponsored box because its relevance score is low.
Why ChatGPT Ads Brand Visibility Matters Now
Most brand visibility work in AI search has, until now, focused entirely on the organic side: structuring content so it gets retrieved, building the entity signals that get a brand named rather than ghost-cited, and tracking share of voice across model responses. That work does not change. What changes is that ChatGPT ads brand visibility is now a second, parallel lever in the same interface, and the two levers reinforce each other through the relevance mechanism both systems share.
Practically, this argues for three adjustments to a brand's AI visibility program:
- Treat entity work as the input to both channels, not only the organic one. The schema, third-party mentions, and consistent brand description that raise organic citation odds (covered in our guide to brand entity optimization for AI search) are the same signals OpenAI's relevance score reads when pricing an ad.
- Do not let ad availability narrow your platform strategy. ChatGPT ads launched in the US, Canada, Australia, and New Zealand, and OpenAI has since added the UK, Japan, South Korea, Brazil, and Mexico (Adweek), but that is still eight countries on one platform. The other major assistants, including Gemini, Claude, and Perplexity, still allocate every mention purely on earned signals, so a ChatGPT-ads-first strategy leaves the rest of the AI search surface uncovered. Our four-stage ChatGPT optimization playbook and the underlying retrieval mechanics it describes still apply regardless of whether you run ads.
- Measure the two channels separately. Ad clicks and impressions are reported in OpenAI's Ads Manager. Organic mentions are not, and most analytics setups undercount them because logged-out ChatGPT sessions carry no referrer data. A brand that only watches ad dashboards will not notice if its organic mention rate is falling.
None of this is reason to delay an ad test if your category fits ChatGPT's commercial-intent queries. It is reason to run that test alongside, not instead of, the entity and content work that earns free placement everywhere else AI assistants generate answers. If you want a baseline read on where your brand currently stands across ChatGPT and the other major assistants before deciding how to split that budget, Elaventra's free AI Visibility Report gives you that starting point.
Frequently asked questions
Do ChatGPT ads affect whether my brand gets cited in the organic answer?
No. OpenAI's ad system is a separate layer that runs after ChatGPT generates its organic response. Buying ads does not change which sources the model cites or which brands it names in the answer text itself. The two systems share a relevance concept but operate independently.
Which AI platforms currently run ads inside chat responses?
As of June 2026, ChatGPT is the only major AI assistant with a live advertising product, available to advertisers in the US, Canada, Australia, New Zealand, the UK, Japan, South Korea, Brazil, and Mexico. Gemini, Claude, Perplexity, and Microsoft Copilot do not currently sell ad placements inside chat responses.
What is a context hint in ChatGPT ads, and how is it different from a keyword?
A context hint is a natural-language description an advertiser sets at the ad group level, describing the conversations where their product is genuinely useful. Unlike a keyword, it is not matched against exact terms; OpenAI's system evaluates it against the live conversation's intent and topic, alongside the user's chat history and past ad interactions.
Does a strong brand entity record lower my ChatGPT ad costs?
It can. OpenAI's auction multiplies your bid by a relevance score before ranking it against competitors, and that score draws on the same kind of contextual clarity that improves organic citation odds. A brand with a clear, consistent entity record and well-targeted context hints is likely to see a higher relevance score and a lower effective cost per click than a brand with vague positioning, even at the same bid.
The Channel Is New, the Fundamentals Are Not
ChatGPT ads give brands a second way to appear in the highest-intent conversations on the platform, and OpenAI's own numbers suggest the format is working well enough to keep expanding. But the underlying lesson from this announcement is not "start advertising in ChatGPT." Managing ChatGPT ads brand visibility well means recognizing that relevance, the same factor that has always governed organic AI citation, now also sets your price in the one AI platform selling placement directly. Brands that have already done the entity and content work will find both channels cheaper and more effective than brands that have not.