ChatGPT Ads in France: What Changes for Brands

Since August 24, 2026, ChatGPT has been showing ads in France to users on the Free and Go plans. Ads appear below the answer, labeled as sponsored, and OpenAI states that they do not influence the answer itself. For a brand, advertising becomes a possible lever, but the organic answer, the one that names brands and competitors, is still where the recommendation happens.

On August 18, 2026, OpenAI announced that ChatGPT Ads would expand to 31 European markets, including France. Here is what these ads are, how they are displayed, what is known about how they work for advertisers, and above all how a brand should weigh buying visibility against earning its place in the answers.

What are ChatGPT Ads?

They are paid ads shown in ChatGPT, below an answer, when a sponsored product or service matches the current conversation. They only concern the free and low-cost plans. Higher paid tiers show no ads.

According to OpenAI's help center, ads may be shown to users on the Free and Go plans. Plus, Pro, Business, Enterprise and Edu subscriptions do not include ads. OpenAI justifies advertising by the cost of infrastructure: it is meant to fund free and low-cost access to ChatGPT.

A free user who does not want ads has two options, again according to OpenAI: upgrade to Plus or Pro, or switch to an ad-free free version with fewer messages and reduced access to some tools such as image generation.

Since when have there been ads in ChatGPT in France?

Since Monday, August 24, 2026. France is one of the 31 European markets that opened that day, six months after testing began in the United States. Self-serve access for advertisers followed on August 31.

The timeline, based on OpenAI's pages:

January 16, 2026: OpenAI publishes its advertising principles and announces upcoming tests in the United States on the Free and Go plans.
February 9, 2026: testing starts in the United States with adult Free and Go users.
March 26, 2026: OpenAI announces the pilot's expansion to Canada, Australia and New Zealand.
August 11, 2026: launch in the United Kingdom, Mexico, Brazil, Japan and South Korea (source).
August 18, 2026: announcement of the expansion to 31 European countries, including Germany, France, Spain, Italy, Sweden, Norway, Denmark, the Netherlands and Austria. OpenAI calls it its largest expansion to date.
August 24, 2026: effective launch in Europe, a date confirmed by French outlet Blog du Modérateur, which also reports that European users had been notified by email on August 15.
August 31, 2026: the self-serve Ads Manager becomes available in the 31 European markets, according to the update to OpenAI's announcement.

How are ads displayed in ChatGPT?

Below the end of the answer, in a separate block. OpenAI says ads are clearly labeled as sponsored and visually distinct from the generated text. The block can contain one or more ads, from one or several advertisers.

What OpenAI's help center specifies:

Placement: ads may appear below the end of an answer when they are relevant to the conversation.
Labeling: they are identified as sponsored and kept separate from ChatGPT's answer.
No endorsement: an ad is a paid placement. Seeing one does not mean OpenAI endorses or recommends the advertiser.
Controls: a menu lets users hide the ad, report it, see why they are shown it, or share it with ChatGPT ("Ask ChatGPT") to ask questions. By default, ChatGPT cannot see the ads being shown.
Exceptions: no ads in Temporary Chats, nor, during the test, in the ChatGPT Atlas browser.

On formats, Blog du Modérateur describes an ad made of the advertiser's name, a favicon, a title, a description, an image and a link. It distinguishes the sponsored answer (brand, short title, description, link), the shopping carousel (product cards with price and availability, powered by a feed) and an image format with a call-to-action button, reportedly in testing. OpenAI does not publish an official list of formats in the pages we reviewed.

How are ads targeted in Europe?

In Europe, ads are not personalized at launch. They are selected based on the topic of the current conversation and limited context such as general location and language. Chat history and memory are not used for European targeting for now.

OpenAI's help center is explicit: personalized ads are initially not available in the European Economic Area or Switzerland. Elsewhere, when personalization is on, OpenAI may use chat history, memory and past interactions with ads. According to Blog du Modérateur, personalization will come later in Europe and will require the user's prior consent.

To choose which ad to show, OpenAI says it combines the context and intent of the conversation, the landing page, the ad copy, and the context hints and targeting provided by the advertiser. When several ads are eligible, relevance and bids decide.

On privacy, OpenAI states that advertisers have no access to conversations, chat history, memories or personal information. They receive aggregated reports, such as the number of views and clicks.

How does ChatGPT Ads work for advertisers?

French advertisers can go through OpenAI's teams and partner agencies or, since August 31, through the self-serve Ads Manager at ads.openai.com. Targeting relies on context hints rather than keywords, with CPM or CPC buying.

What OpenAI confirms in its European announcement:

Access: first through OpenAI's Ads Solutions team, partner agencies and technology partners, then self-serve through Ads Manager.
Buying: beyond CPM and CPC bidding, the platform supports conversion optimization.
Advertiser targeting: geographic targeting and custom audiences have been added.
Measurement: the OpenAI Pixel, a Conversions API and integrations with third-party measurement solutions.
Adoption: OpenAI says tens of thousands of marketers have already run ads on ChatGPT.

Blog du Modérateur lists the six agency groups partnering on the European launch: Publicis, WPP, Omnicom, MediaPlus, Havas and dentsu. In its practical guide, it details how Ads Manager works: a campaign, ad group and ad structure, "context hints" that describe the conversations you want to reach, no minimum budget, and very short creatives (16 characters for the title, 32 for the description). It also notes that landing pages must not block the OAI-AdsBot and OAI-SearchBot crawlers. According to the same outlet, the minimum commitment required at the start of the pilot has been gradually removed.

What OpenAI does not publish in the pages we read: a rate card, benchmark CPMs or CPCs for France, or average click-through rates. The performance figures in circulation (lower acquisition costs, the share of conversations with commercial intent) come from OpenAI press statements, without a detailed methodology.

Before buying visibility, check whether your site is ready to be cited by AI. Free GEO score on 7 criteria, with a citation test on ChatGPT.

Analyze my site for free →

What are the rules and limits?

No ads for identified minors, no ads next to sensitive topics like health, mental health or politics, and no political advertising. Above all, OpenAI states that advertising has no effect on the organic answer.

The rules published by OpenAI (help center, test announcement):

Minors: no ads for accounts identified as belonging to users under 18, based on declared age and, where applicable, age prediction.
Sensitive topics: no ads near conversations about personal health, mental health or politics. Some advertisers in regulated sectors (health, financial services) may be admitted under strict criteria, but never next to sensitive conversations.
Political advertising: currently not allowed on ChatGPT.
Answer independence: according to OpenAI, ads run on systems separate from the chat model, and advertisers cannot influence, rank or change answers.
Data: ad data deleted by the user is removed from servers within 30 days.

These safeguards are commitments made by OpenAI. They cannot be verified from the outside at this stage. On the European side, Blog du Modérateur notes that the data controller is OpenAI Ireland Limited and that French users can file complaints with the CNIL, France's data protection authority.

Advertising or organic visibility: what matters most for a brand?

The organic answer. It names the recommended brands, compares options and cites your competitors. Ads appear below it, labeled as sponsored, and cannot change what the answer says about you. The two complement each other, but one does not replace the other.

Take a typical question: "which payroll software should a 30-person SME choose?". ChatGPT answers by naming three or four solutions, with their strengths and weaknesses. Below the answer, a sponsored block may present another offer. Three takeaways:

Trust is not the same. The answer is perceived as advice. The ad is explicitly a paid placement, and OpenAI itself says it is not a recommendation.
Your competitors are in the answer itself. If ChatGPT names three competitors and not you, your ad comes after a recommendation that leaves you out. We covered this mechanism in our article on competitors cited by ChatGPT.
Ads only reach part of the audience. Plus, Pro, Business, Enterprise and Edu subscribers see no ads, nor do free users who switch to the ad-free version. The organic answer is seen by everyone.

Advertising remains useful in several cases: launching a product that AI engines do not know yet, occupying a category where you are rarely cited while you build authority, or pushing a specific commercial offer. But it does not fix what ChatGPT says about your brand. If the answer is inaccurate or unfavorable, that is an AI reputation issue, not a bidding one.

That is why the first step is not opening an ad account, but knowing where you stand. Beeleven's AI visibility audit, from the agency that built Detekia, asks ChatGPT, Gemini, Claude and Perplexity the questions your prospects ask, measures whether your brand is cited, in which position and against which competitors, and weights the result by the real usage of each AI engine in France. You then know which questions already recommend you, and where an ad would have a real role to play.

→How ChatGPT chooses its sources (and how to be one of them)

How do you combine advertising and GEO?

Start from organic measurement. Where ChatGPT already cites you, ads are not a priority. Where it cites your competitors, ads can act as a temporary bridge while you work on GEO. And where the answer is wrong, you need to fix the sources, not buy impressions.

A simple grid, question by question:

Situation in the organic answer	Role of advertising	GEO priority
You are cited prominently	Low, except for one-off offers	Maintain: up-to-date content, external sources
You are cited, but after competitors	Possible complement on high-intent queries	Strengthen proof, reviews, comparisons
You are absent, competitors are cited	Temporary bridge to be part of the conversation	High: understand why they are chosen
The answer is inaccurate about your brand	Useless for correcting it	Urgent: fix the cited sources

For e-commerce, the shopping carousel described by Blog du Modérateur relies on a product feed: a clean feed serves both advertising and organic product recommendations. For software vendors, the logic is the same as getting recommended by ChatGPT as a B2B SaaS: an ad can get you into a comparison, but only the answer puts you on the shortlist.

Finally, keep ChatGPT's weight in mind. According to Arcom's barometer, it accounts for 76% of time spent on AI services in France. We analyze these figures in our article on Arcom's AI barometer. An ad on ChatGPT reaches a lot of people, but so does being absent from its answers.

What should you do in practice?

Measure your organic visibility on your customers' questions first, fix what is blocking you, then test advertising where it has a specific role. Six steps let you decide without spending blindly.

List your prospects' questions. 20 to 50 real questions: comparisons, "best X for Y", problems you solve. These are the conversations where an ad could appear.
Measure what ChatGPT answers. Are you cited? In which position? Who is cited instead of you? Methods are compared in our guide to measuring AI visibility. For a full measurement across four AI engines, weighted by usage, see the AI visibility audit.
Check that your site is readable by OpenAI's crawlers. OAI-SearchBot for search, and OAI-AdsBot if you advertise. A site that blocks AI crawlers penalizes itself on both sides.
Work on the queries where you are absent. Direct answers at the top of the page, sourced figures and proof, presence on the third-party sources ChatGPT consults.
Test advertising with a precise goal. One product, one category, one period. Small budget, a few weeks, and conversion tracking in place before launch.
Measure over time. Answers change, and so do ad rules, especially once personalization arrives in Europe. Re-run your organic measurement regularly to make sure ads are not hiding a decline in the answers.

Step one: check for free whether your site is ready for AI. GEO score on 7 criteria and a citation test on ChatGPT, in under a minute.

Analyze my site for free →

Frequently asked questions

Are ChatGPT ads available in France?

Yes. They have been shown since August 24, 2026 to Free and Go users, as part of the expansion to 31 European markets. For advertisers, the self-serve Ads Manager has been open in France since August 31, 2026.

Can a brand pay to be cited in ChatGPT's answer?

No, according to OpenAI. Ads appear below the answer, labeled as sponsored, and advertisers cannot influence, rank or change answers (OpenAI help center). Being cited in the answer is a matter of organic visibility, in other words GEO.

Are ChatGPT ads personalized in France?

Not at launch. In the European Economic Area and Switzerland, ads rely on the current conversation and limited context. Personalization, which exists elsewhere, is initially not available in Europe.

Should you advertise on ChatGPT or work on GEO?

Start by measuring your organic visibility. If ChatGPT already cites you, ads are secondary. If competitors own the answer, ads can help in the short term, but GEO remains the only way into the recommendation itself. Beeleven's AI visibility audit is designed to make exactly this diagnosis.

Sources

→Zero-click: when AI answers without sending traffic

→GEO audit: how to audit your AI visibility