ChatGPT advertising went live across 31 European markets today. If you run marketing for a business of any size, you will hear about it this week — probably from someone offering to get you into ChatGPT.
Before anyone spends anything, it is worth being precise about what has actually launched, because the gap between what this is and what it will be sold as is unusually wide.
What happened, in order
The rollout has been methodical rather than sudden, which is easy to miss if today is the first you have heard of it.
- February 2026 — OpenAI began testing ads in ChatGPT with a pilot in the United States.
- March 2026 — the pilot extended to Canada, Australia and New Zealand.
- May 2026 — OpenAI signalled expansion into the United Kingdom, Mexico, Brazil, Japan and South Korea.
- 11 August 2026 — those markets went live. The UK has had ChatGPT advertising for a fortnight.
- 18 August 2026 — OpenAI announced the European expansion.
- 24 August 2026 — today — 31 European markets go live, including Germany, France, Spain, Italy, the Netherlands, Austria, Poland, Sweden, Norway, Denmark and Ireland.
That last step takes the platform from a pilot in a handful of countries to something close to continental coverage in one move.
For UK businesses there is a small but useful point buried in that timeline. This is not new to you. It has been available in your market since the 11th, and while attention has been on the European announcement, the channel has quietly been open.
The three things being glossed over
1. Advertising does not buy you a place in the answer
This is the one that matters most, and it is the one most likely to be blurred in a sales conversation.
OpenAI states that ads do not influence the answers ChatGPT gives. Advertising appears as a separate, clearly marked placement running alongside the response. It does not purchase a mention inside the organic answer, it does not make the assistant more likely to recommend you, and conversations are kept private from advertisers.
That distinction is the whole ballgame. When someone asks ChatGPT which agency, supplier or product to choose, the answer they get is assembled from what the model considers credible published information. You cannot buy into it. You earn your way into it, through coverage, consistency and authority — which is to say, through public relations.
So if a proposal implies that spending on ChatGPT ads will improve how the assistant talks about you, that proposal is describing a mechanism that does not exist. It is a reasonable first question to ask anyone pitching you this week.
2. In the EEA, this is contextual advertising, not audience targeting
The capability list circulating around this launch includes custom audiences. For much of Europe, that is not the position at launch.
OpenAI has said personalised ads are not initially available in the European Economic Area or Switzerland. Where ads do appear in those markets, selection draws on the current conversation topic, approximate location, device type, time of day and language. Past conversations and stored memories are not used.
Read plainly: European ChatGPT advertising begins as contextual advertising. It responds to what someone is asking about right now, not to a profile built from what they have asked before.
That is not a weakness — contextual placement against genuine purchase intent is a respectable proposition, and arguably a cleaner one. But it changes the plan. Strategies built on audience segments and retargeting logic do not transfer. Strategies built on being relevant to a live question do.
3. The reachable audience is narrower than “everyone who uses ChatGPT”
Ads are shown to logged-in adult users on the Free and Go plans. Plus, Pro, Business, Enterprise and Education accounts do not see them.
If your buyers are senior people whose employers pay for ChatGPT at work, a meaningful share of them are sitting in the ad-free tiers. Teen accounts are excluded too, following the launch of ChatGPT for Teens the same week.
None of this makes the channel unattractive. It makes it a channel with a shape, and the shape should inform the budget.
What this changes for how you measure
The measurement risk here is specific and predictable. Organisations will run ChatGPT ads, notice that AI assistants seem to mention them more often, and conclude the advertising caused it.
Usually it will not have. What tends to move organic answer presence is coverage, published clarity and citation — the earned side. If a paid campaign happens to run at the same time as a PR push, the paid line item will get the credit unless you separate them deliberately.
We keep four layers apart, each with its own scorecard:
| Layer | What it is | Measured on |
|---|---|---|
| Organic AI answer | Whether the assistant names and recommends you unprompted | Mentions, wording, citations, recommendation frequency |
| ChatGPT advertising | Marked placement running alongside the answer | Impressions, engagement, conversions |
| Earned authority | Coverage and validation in sources the models trust | Coverage quality, source authority, subsequent answer changes |
| Commercial result | What the business actually receives | Qualified enquiries, pipeline, revenue |
One rule holds across all of it: paid advertising does not buy influence over organic AI answers. Where the two move together, say so and show why. Do not let a campaign claim ground that was earned.
What we would do this week
If you are considering this, three practical steps before any spend.
Establish where you stand organically first. Run the questions your buyers actually ask across ChatGPT and the other assistants, and record what comes back — whether you are named, how you are described, who is recommended instead. Without that baseline you have no way of telling later whether anything changed, or why.
Treat the paid layer as contextual. In the EEA, plan around the questions people are asking rather than the audiences you wish you could target. Ask what someone is trying to solve at the moment your placement appears.
Fix the record before you amplify it. If assistants are describing your business from outdated or inconsistent sources, advertising alongside that answer does not correct it. The published material has to change first.
Today is a genuine shift: ChatGPT is becoming a performance channel with real bidding, real geographic targeting and real measurement infrastructure. It deserves serious consideration.
It also deserves the accuracy the moment warrants. Being present in the answer and being present beside the answer are two different disciplines. One is bought. The other is earned — and the second is where reputations are actually made.
3RD4PR helps organisations measure and improve how AI assistants describe them, and runs the paid layer alongside it with the two kept honestly apart. If you want to know what ChatGPT says about your business today, submit a brief — or read more about AI Search & Discovery.
Common questions
Can you pay to appear in ChatGPT's answers?
No. OpenAI states that advertising does not influence the answers ChatGPT gives. ChatGPT Ads is a separate, marked placement that appears alongside the answer. It does not buy a mention inside the organic response, and any agency implying otherwise is describing something that does not exist.
Which European countries can advertise on ChatGPT now?
ChatGPT Ads expanded to 31 European markets on 24 August 2026, including Germany, France, Spain, Italy, the Netherlands, Austria, Poland, Sweden, Norway, Denmark and Ireland. The United Kingdom was part of an earlier wave that went live on 11 August 2026, alongside Mexico, Brazil, Japan and South Korea. Before that, the pilot ran in the United States from February 2026 and in Canada, Australia and New Zealand from March.
Are ChatGPT ads personalised in Europe?
Not at launch. OpenAI has said personalised ads are not initially available in the European Economic Area or Switzerland. Ad selection there draws on the current conversation topic, approximate location, device type, time of day and language — not on past chats or stored memories. That makes European ChatGPT advertising contextual rather than audience-based for now.
Who actually sees ads in ChatGPT?
Ads are shown to logged-in adult users on the Free and Go plans. Plus, Pro, Business, Enterprise and Education accounts do not see them, and teen accounts are excluded. That matters for planning: the reachable audience is the free and low-cost tier, not the paying professional user base.
How do I measure ChatGPT advertising against organic AI visibility?
Separately, and deliberately so. Paid placement is measured on impressions, engagement and conversions. Organic answer presence is measured on whether the assistant mentions you, how it describes you, which sources it cites and how often it recommends you. Collapsing the two is the most common error we expect to see, because it lets a paid campaign take credit for authority that was earned elsewhere.
How do advertisers buy ChatGPT ads?
Initially through OpenAI's Ads Solutions team, agency partners and technology partners. OpenAI has said self-service buying through an Ads Manager will follow later in the summer. Measurement is supported through the OpenAI Pixel and Conversions API, with third-party measurement integrations available.