ChatGPT Ads in Europe: What It Means for Your Stack

OpenAI announced on 18 August that ChatGPT Ads is expanding to 31 European markets the following week, six months after it began testing ads in the United States. It is the company's largest ad expansion so far.

For most European marketing teams the ad placements themselves are the least interesting part of this. A new inventory source appears roughly once a quarter and you evaluate it on cost and performance like any other. The part worth an hour of your attention is the measurement stack that comes with it: the OpenAI Pixel, the Conversions API, and the third-party integrations built around them. That is where a US advertising platform stops being something you buy from and starts being a processor inside your own site and your own server-side data flow.

What Launched

Ads appear in ChatGPT for users on the Free and Go plans across 31 European markets. OpenAI names nine of them: Germany, France, Spain, Italy, Sweden, Norway, Denmark, the Netherlands and Austria. It has not published the full list, which is worth knowing if you are planning campaign geography. Note also that the footprint is European rather than strictly EU, since Norway is EEA rather than a member state.

Paid tiers are unaffected. Plus, Pro and Enterprise remain ad-free, which matters more than it sounds: the audience you are reaching through this channel is by definition the segment that has not paid for ChatGPT.

On the buying side, OpenAI has moved past simple CPM and CPC bidding to conversion optimisation, added geo-targeting and custom audiences, and extended measurement beyond clicks through the OpenAI Pixel, a Conversions API and third-party measurement integrations.

Access is gated at launch. Advertisers buy through OpenAI's Ads Solutions team, agency partners or technology partners, with self-service access through Ads Manager arriving later. If you are a small in-house team, you are probably waiting.

What OpenAI Has Published, and What It Has Not

This distinction is the most useful thing in the announcement, and it is easy to miss because most coverage has blurred it.

The announcement is a go-to-market document. It does not state a legal basis for ad targeting in the EU. No mention of consent, legitimate interest, or how the GDPR applies. What OpenAI does commit to, there and in the ads principles it links to, is this:

Conversations are kept private from advertisers, and customer data is never sold to advertisers
Ads are clearly labelled, kept separate from answers, and do not influence the answers ChatGPT gives
Users control ad personalisation, can clear the data used for ads at any time, and can pay for an ad-free tier
During the US pilot, ads were based on the current conversation, were withheld from accounts predicted to belong to under-18s, and were not eligible near sensitive or regulated topics such as health, mental health and politics

Press coverage of the European launch goes considerably further, describing a two-tier design in which contextual ads run under legitimate interest while personalised targeting based on ad history requires opt-in, with targeting signals limited to the current conversation, approximate location, device type, time of day and language, and chat history and saved memories excluded.

That may well be accurate. OpenAI has not published it, so treat it as reporting rather than as a commitment you can put in front of your data protection officer.

The distinction is not pedantry. "We never sell your data to advertisers" and "personalised targeting is off unless you opt in" are different promises with different consequences, and only the first is in writing from the company. If the second matters to your risk assessment, ask OpenAI to confirm it in the contract rather than citing a trade publication.

What Actually Lands on You

Here is the part that changes your compliance posture rather than OpenAI's.

If you adopt the OpenAI Pixel or the Conversions API, you are adding a US advertising technology provider to your own processing chain. This is the same category of decision as the Meta Pixel or Google Ads conversion tracking, and it raises the same three questions, none of which are answered by OpenAI's consent model for ChatGPT users:

Consent on your side. A pixel that writes to or reads from the user's device needs consent under the ePrivacy Directive, and that is your consent banner, not OpenAI's. This is independent of whether the ad that drove the click was contextual or personalised. See what changed in cookie consent in 2026.
A DPA and a transfer basis. You need a Data Processing Agreement covering this processing, and you need to know which mechanism carries the transfer: certification under the EU-US Data Privacy Framework, Standard Contractual Clauses, or something else. Ask for the document rather than assuming. The framework's status is not static, and a transfer basis that is fine today is a risk register entry tomorrow.
What the Conversions API actually sends. Server-side conversion tracking is attractive precisely because it bypasses browser restrictions, which means it is also easy to send more than you intended. Hashed emails and order values are personal data. Decide what leaves your servers before you wire it up, not after. Our Schrems II compliance checklist covers the questions worth asking.

None of this makes ChatGPT Ads unusable in Europe. It makes it a procurement decision rather than a checkbox, and one that belongs with whoever owns your data protection documentation rather than solely with whoever owns the media budget.

Paid Placement and Organic Visibility Are Now Separate Problems

Until now, "being visible in ChatGPT" meant one thing for European brands: whether the model cited you when someone asked a question in your category. That is the discipline the European GEO software category exists to measure, and it remains the larger opportunity for most businesses because it costs media budget nothing.

Now there are two distinct surfaces. Organic citation in the answer, and paid placement alongside it. They are bought, measured and optimised differently, and a tool that tracks one does not track the other. If you are already running AI visibility monitoring, do not assume your existing dashboard covers this.

One honest caveat about performance: this has only just launched. Anyone offering you European benchmarks for ChatGPT ad performance right now is extrapolating from a six-month US test in a different regulatory and linguistic market. Treat early case studies accordingly.

What To Do This Week

If ChatGPT Ads are on your roadmap:

Establish whether you can even buy yet. Self-service is not live, so this runs through OpenAI's team, an agency partner or a technology partner.
Get the DPA and the sub-processor list before the pixel goes on the site, not during the first campaign.
Decide what the Conversions API is permitted to send, and write it down.
Check your consent banner actually gates the new pixel. A tag added outside your consent management platform is the most common way a compliant setup quietly stops being one.
Keep your AI visibility tracking separate from your paid reporting. They answer different questions.

If they are not on your roadmap, the practical takeaway is smaller: nothing about this changes your existing stack until you install something.

Frequently Asked Questions

Do ChatGPT ads in Europe use my chat history for targeting?

OpenAI's published position is that conversations are kept private from advertisers and never sold, and that during its US pilot ads were based on the current conversation. Press reporting on the European launch says chat history and saved memories are excluded from targeting, but OpenAI has not published that itself.

Can I turn ChatGPT ads off?

OpenAI says users can turn off personalisation and clear the data used for ads, and that there will always be a way not to see ads, naming the ad-free paid tiers. On the Free and Go plans, the personalisation control governs which ads you see rather than whether ads appear. Plus, Pro, Business and Enterprise plans are ad-free.

Does using ChatGPT Ads make my business non-compliant with GDPR?

No. It makes you responsible for a new processing relationship. You need a DPA, a documented transfer basis, consent for any pixel that touches the user's device, and a decision about what your server-side tracking sends. Those are the same obligations that come with any US advertising platform.

Which European countries are included?

OpenAI says 31 markets but names only nine: Germany, France, Spain, Italy, Sweden, Norway, Denmark, the Netherlands and Austria. The full list has not been published. Norway's inclusion confirms the footprint is European rather than EU-only.

Is there a European alternative to advertising in ChatGPT?

Not a like-for-like one, because the inventory is specific to ChatGPT. The closer question is whether paid placement is the right instrument at all. For most European businesses, earning citations in AI answers is still the better-value channel, and European GEO tools measure it.

If you are reviewing what else sits in your stack while you have the data protection documentation open, our directory lists European marketing software with the hosting location, ownership and DPA status verified against each vendor's own published documents.