We earn a commission if you buy through some links on this page. It doesn’t change our rankings or the price you pay.
Yes, ChatGPT shows ads on the Free and Go plans in many countries. Here’s where things stand as of 27 October 2026.
On 21 October, OpenAI expanded ads to 26 more countries, for a total of 89. That list includes the US, Canada, Australia, the UK, Japan, Brazil, India and the whole EU. Plus, Pro, Business, Enterprise and Edu stay ad-free. The ads sit below the answer with a “Sponsored” label.
Can you block them? Browser users report mixed results. EasyList, uBlock filters and AdGuard Base all carry rules aimed at ChatGPT ads, which we checked on 27 October 2026. The fix history is messy, though. We haven’t tested blocking ChatGPT ads ourselves, so treat everything in the ad blocker section as evidence, not a promise.
OpenAI’s help pages wouldn’t load for us, so some OpenAI details below come from press coverage and search snippets. We flag it where it matters.
Does ChatGPT Have Ads? The Rollout So Far
It started as a US test and turned into a global rollout in about nine months.
| Date | What happened |
| 16 Jan 2026 | OpenAI announces plans to test ads for US Free and Go users |
| 9 Feb 2026 | US pilot starts for logged-in adults |
| Mar to Apr 2026 | Canada, Australia and New Zealand added |
| Jun to Aug 2026 | UK, Mexico, Brazil, Japan and South Korea; OpenAI confirmed all five as launched on 11 Aug |
| Aug 2026 | India added |
| 24 Aug 2026 | 31 European markets go live: the EU-27 plus Iceland, Liechtenstein, Norway and Switzerland |
| 5 Oct 2026 | Visual ads next to image-generation results announced, US only |
| 21 Oct 2026 | 26 more countries, 89 in total |
Some dates are fuzzy. Japanese and Korean press reported ads live by late June, weeks before OpenAI’s own “now launched” update. Brazilian outlets disagree on whether the announcement came on 4 or 7 August.
OpenAI also says more than 100,000 advertisers have used the platform. With that many buyers, we wouldn’t bet on ads going away.
Who Sees Ads in ChatGPT?
Your plan, country, age classification, platform and chat type affect whether you see ads.
By Plan
| Plan | Ads? | US price (Jan 2026) |
| Free | Yes | $0 |
| Go | Yes | $8/month |
| Plus | No | $20/month |
| Pro, Business, Enterprise, Edu | No | Not covered here |
Go is the odd one out. You pay $8/month in the US and still get ads. Prices are localised elsewhere, for example around ¥1,500/month in Japan through the app. Our price figures date from January, so check the live price in your country before you pay.
By Country or Region
The rollout covers 89 countries. The October batch named 14 Latin American countries, including Argentina, Chile, Colombia and Peru, plus Nigeria, South Africa, Serbia, Ukraine, Cambodia, Bangladesh and Pakistan. OpenAI also has a post about Southeast Asia and Taiwan, but we couldn’t read its date or country list.
We couldn’t get the full list of 89.
Europe gets a slightly different version. At launch, ads in the EEA and Switzerland used only the current conversation, rough location and language, because OpenAI said personalised ads wouldn’t be available there “initially”. That word leaves room for change.
By Platform
Ads show up on the web and in both mobile apps, and since August advertisers can choose iOS, Android or Web, where Web covers desktop and mobile browsers. Our sources don’t confirm whether the macOS and Windows desktop apps show them.
There were no ads in OpenAI’s Atlas browser during the test.
What ChatGPT Ads Look Like
A ChatGPT ad is a separate box under the reply, labelled “Sponsored” and set apart from the answer. German users have reported the label “Anzeige”. Brazilian coverage describes the same spot: the end of the screen, right after the answer.
Each unit can carry an advertiser name, favicon, headline, description, landing URL and image. OpenAI recommends headlines of about 16 characters and descriptions of about 32. That’s tiny, shorter than most sentences on this page.
An options menu on the ad tells you who the sponsor is and why you’re seeing it. A newer format puts visual ads next to generated images, announced for the US in October. We don’t know whether that format is live yet.
The answer itself isn’t an ad. OpenAI says ads don’t influence what ChatGPT tells you. That’s OpenAI’s promise, and you can’t check it from the outside.
How ChatGPT Ads Are Targeted
The basic signal is the topic of your current conversation. Outside the EEA and Switzerland, ChatGPT can also use your ad history, interests and past chats or memory, unless you switch that off.
Guides describe the switches under Settings > Ads controls, with a “Personalize ads” toggle and a separate one for past chats and memory. Labels may differ slightly in your app. You can delete your ads history and interests there too. Deleted ads data can stay on OpenAI’s servers for up to 30 days.
Advertisers don’t pick keywords or behavioural profiles, but instead supply “context hints” for each ad group and pay either per thousand views or per click. In return they get aggregated views and clicks. Your chats, memories, name, email, precise location and IP address aren’t shared with them.
Where ChatGPT Doesn’t Show Ads
- Under 18s: accounts that are, or are predicted to be, under 18 get no ads. That relies on OpenAI’s age prediction, which rolled out worldwide on 18 August 2026 and can get it wrong in both directions.
- Sensitive topics: no ads near physical health, mental health or politics, and political ads aren’t allowed for now.
- Temporary Chats: no ads.
How to Get Rid of ChatGPT Ads
1. Pay for Plus or Higher
Plus and every plan above it are ad-free. Go isn’t, so Go users have two ways out: upgrade to Plus, or drop back to Free and pick its ad-free mode with reduced limits. Plus was $20/month in the US when Go launched in January, so check the current price before you pay.
If you already wanted Plus for the features, this is the cleanest fix. Paying $20/month just to hide a sponsored box is a harder sell.
2. Switch Free to the Ad-Free Mode
Free users can turn ads off without paying. OpenAI’s help page, as quoted in the press, puts it under Settings > Ads controls > “Change plan to go ad-free”, then “Reduce message limits”.
The catch? Fewer daily messages. You also lose features, and OpenAI names image generation and deep research as gone. When you hit the limit, a banner offers an upgrade or a return to the ad-supported version. We didn’t find the exact message limits published.
If you need image generation or deep research, this mode won’t suit you.
3. Use Temporary Chats
Temporary Chats don’t show ads, according to OpenAI’s FAQ as reported by Notebookcheck. It’s a per-conversation workaround rather than a setting, but it costs nothing for one-off questions.
Turning off ad personalisation won’t help, because it only changes how ads are picked and doesn’t remove them.
Can You Block ChatGPT Ads With an Ad Blocker?
Sometimes, on the web. Here’s what that’s based on.
What we actually checked were the filter lists themselves and the public bug reports on GitHub, which we read on 27 October 2026. We didn’t run a blocker against ChatGPT, so none of this is a test result.
The Filter Lists Have Rules
- EasyList: contains rules targeting ChatGPT’s “bazaar” ad endpoint and “message_supplements” requests, plus about 13 cosmetic rules intended to hide ad boxes. EasyList is on by default in uBlock Origin, uBlock Origin Lite and AdGuard.
- uBlock filters: scriptlets that edit ChatGPT’s response stream, marking ads as hidden and stripping ad fields before they render. These need full uBlock Origin; whether uBlock Origin Lite applies them is unverified.
- AdGuard Base, uBlock Origin version: similar response-pruning rules plus rules for image cards with “Sponsored options” and “Ad options” buttons. The native AdGuard build may differ. Those may target the new visual ads, but that isn’t confirmed.
The Bug Reports Show a Cat-and-Mouse Game
The main uBlock Origin issue opened on 16 March 2026, and a cosmetic rule fixed it two days later. In August the ads came back. A maintainer noted that the new rule removed its target but ads still appeared.
On 10 October a new rule went in, and one tester reported asking 10 questions without seeing an ad. Six days later, maintainers disabled a change “as causing freeze”. The issue is still open, labelled “geo specific” and “ongoing issue”.
uBlock Origin Lite users on Chromium reported leftover ad remnants in September. On the AdGuard side, one Windows user saw the ad removed but a “learn more” block left behind. AdGuard closed its latest ChatGPT report on 26 October as “Not reproduced”. Not reproduced isn’t the same as fixed, especially when ads vary by region.
Even AdGuard was cautious early on. In February it expected web ads to be blockable but called app blocking “much more tricky”.
The Mobile Apps Are an Open Question
We found no evidence that browser filters affect the ChatGPT iOS or Android apps. We also found no reports, good or bad, on system-wide blockers there.
The web rules target ad requests on chatgpt.com itself, not a separate ad domain. So DNS blocking probably can’t single them out without blocking ChatGPT too, though nobody has documented how the apps fetch their ads. For other apps, our guide to blocking in-app ads on Android and iPhone covers what usually works.
Which Ad Blocker to Try
Two blockers have active ChatGPT rules and public bug threads: uBlock Origin and AdGuard.
uBlock Origin on Firefox is free and supports the scriptlet rules described above. Its bug thread is also the messiest, with fixes, regressions and one rule pulled for freezing the page. Firefox is still where full uBO runs, as our page on whether uBlock Origin works on Firefox explains. On Chrome, uBlock Origin Lite gets EasyList but had leftovers in September. Our uBlock Origin vs uBlock Origin Lite comparison covers why the Lite version can’t do everything.
AdGuard has ChatGPT rules in AdGuard Base, including the image-card rules, though we only read the uBlock-format version of that list. Its latest ChatGPT report closed on 26 October as “Not reproduced”. That makes it a reasonable pick if you already want one blocker across browsers and apps. Just not for ChatGPT’s mobile app. We found no evidence either way there. For a side-by-side view, see AdGuard vs uBlock Origin.
Skip the “ChatGPT ad blocker” extensions that quote block rates without a method. Also skip one-rule community lists built on hashed class names, which break whenever the interface changes.
If an ad gets through, update your filter lists first. If it’s still there, report it on the filter list’s GitHub tracker with your country and browser, because the “geo specific” label suggests location matters. Our explainer on why ad blockers can’t block every ad goes into the general reasons.