ChatGPT ads now run in dozens of countries. Should you buy them, and do they touch organic answers?

What OpenAI announced on 23 September

OpenAI said on 23 September 2026 that ChatGPT Ads were rolling out across Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam and Taiwan. With that step, it said, ads are available in over 60 countries.¹ Advertisers can buy through OpenAI's sales team, agency and technology partners, or self-service in Ads Manager for eligible businesses.

Ads appear only for users on the Free and Go plans; Plus, Pro and Enterprise stay ad-free.¹ The same announcement repeats OpenAI's statement from the end of August that ChatGPT Ads reached $1 billion in annualised revenue run rate less than 200 days after launch, and says tens of thousands of advertisers now run ads.¹ That is OpenAI's own figure, and the company sells the product.

This note looks at a narrower question: what buying ads does, and does not do, to the organic answer about your brand.

Do ads touch organic answers?

OpenAI says no. Its ads page, first published in February 2026 and updated since, states that ads do not influence the answers ChatGPT gives, that they are clearly labelled as sponsored and visually separated from the organic answer.² It also says advertisers do not see users' chats, and that ads are matched to the conversation topic, past chats and past ad interactions.²

Shopping results sit under a separate statement. Search Engine Journal's September report on ChatGPT Shopping quotes OpenAI saying product results are not ads and are not influenced by any OpenAI partnership, and covers how those results now lean on merchant product feeds.³ Our note on the feed shift covers what to check there.

Our view: take the separation at its word for planning, and verify it with measurement. Nobody outside OpenAI can audit the ranking. What you can do is track your organic share of answer for the same prompts before, during and after a campaign. If it moves only while you pay, you will see it.

How dense are ads already?

Dense enough to matter for a buyer's attention. Similarweb reports that roughly 26% of ChatGPT responses now contain an ad (sample, period and geography not stated).⁴ In the same report, about a third of ad impressions appear in the very first response, and about a fifth come ten or more messages into a conversation (period not stated).⁴ Similarweb sells digital market data.

Links in organic answers remain scarcer than that. In Similarweb's data on US ChatGPT prompts, the share carrying a citation rose from about 1.6% in June 2025 to about 6.8% by May 2026.⁵ Our note on citation rates by sector shows how much that varies.

The comparison is loose, because the two figures come from different samples and periods. The direction is still clear: on ChatGPT, a paid placement may now be a likelier source of a clickable link than an organic citation. That is a reason to consider a test, and no reason to stop organic work, because the answer text itself still decides whether you are named.

Should you buy them?

Buy them as a test if your buyers use ChatGPT's free tiers, your category is commercial and you can measure what happens after the click. Skip them for now if your buyers are mostly on paid plans at work, where ads do not appear, or if your sales cycle is too long to attribute a click.

Treat the budget like any new channel. Our survey of published GEO prices shows what organic AI visibility work costs; an ad test is a separate line, not a replacement. The discipline in our guide to comparing agent pricing applies here too: convert spend to a cost per outcome you actually value, such as a qualified lead.

A small ChatGPT ads test

1. Baseline

   Measure organic share of answer for your buyer prompts for two weeks.

2. Scope

   One country, one product line, a fixed budget and a fixed end date.

3. Track

   Tag landing pages so ad clicks are separate from organic AI referrals.

4. Compare

   Cost per qualified lead against search ads; organic share during and after.

5. Decide

   Scale, pause or stop, with the numbers written down.

What to watch

Watch three things as ads spread. First, whether OpenAI's separation statement changes in its ads principles or help pages; we would re-read them each quarter. Second, ad density in your own category, by recording when ads appear beside your buyer prompts. Third, your organic share on ChatGPT against engines where you do not advertise, so any difference shows up.

Our view: ads buy a place next to the answer, never inside it. The answer still depends on what ChatGPT finds about you: your own pages, third-party profiles and the facts they agree on. Our AI answer monitor tracks that organic share across six engines, using official APIs where offered and a licensed AI-answer data provider otherwise.