October 21, 2026 · Daily Radar
On October 20, OpenAI sent an email to advertisers using ChatGPT Ads, and the most important sentence sits right in the middle: "When enabled, these ads can run automatically without requiring you to review or approve each version." In other words, a headline can change, a description can be rewritten, and you may never have seen that version.
This is less a new feature announcement than a change to the default setting. As of Friday, October 30, 2026, "creative enhancements" will switch on automatically in both new and existing self-serve campaigns. You have nine days to turn them off. If you don't, you can still turn them off later, but until then nobody will show you what was served.
Quick answer: Starting October 30, 2026, OpenAI is turning creative enhancements on by default in ChatGPT Ads; the headline, description, image size, translation and destination URL can change without the advertiser's approval. It affects anyone running ads in ChatGPT or thinking about it. What to do: open the Ads Manager setting by October 29 and make a deliberate decision about whether to leave it on or turn it off.
Search Engine Roundtable broke the news on October 20, quoting the email OpenAI sent to advertisers [A]. The email's main point: as of Friday, October 30, 2026, creative enhancements will be enabled in new and existing self-serve campaigns. Search Engine Land, PPC Land and Germany's Dr. Web reported the same change with the same date [B].
What the system can touch is spelled out too. According to OpenAI's help page, working from your existing ad and landing page content, it can generate new or updated headlines and descriptions, resize images, translate ad copy and "optimize" the destination URL [B]. Variations can be prepared in advance or generated on the spot based on the conversation where the ad is shown.
One more detail: versions generated in real time don't sit in a list you can review before they go live. Under section three of the Ad Tools Terms, the advertiser is responsible for these versions, including claims about price, stock, performance and expected results. OpenAI doesn't guarantee that the versions are accurate [B].
The way to turn it off is clear. It's done in the Ads Manager settings, at the account level or per campaign; the campaign setting overrides the account setting, and opt-outs you made earlier are kept. Sources give the last day as October 29 or 30. I read it as "before the 30th," and to be safe I'm aiming for the 29th.
To me the issue isn't that ad copy gets written with AI; it's that the approval step for that work disappears by default. The output is the same, but the "read it and put it live" step is no longer in your hands, and if you forget to turn it off, it doesn't exist.
I see risk in three places. If the destination URL changes, the page in your measurement may not be the page you designed. If copy is written on the spot based on the conversation, an inappropriate phrase could come out in a regulated industry like health, finance or law, and go live without anyone seeing it. If translation is on, it's theoretically possible for an ad you targeted in Turkish to be shown in another language; I don't know how often that will happen in Turkey, and I'm not speculating.
The advertiser reaction is out there too. Melissa Mackey wrote that she paused her ChatGPT ads test because of poor lead quality. Anthony Higman says that the defaults platforms offer as "less work for you" usually mean impressions in places you didn't want [C]. Both are individual user comments, not general data. Still, at first glance they don't sound unreasonable.
In Turkey: does this apply? In September we wrote that ChatGPT's ad dashboard had opened in Turkey, but without lira support: ChatGPT Ad Dashboard Opens in Turkey: No Lira. Whether creative enhancements have reached accounts in Turkey too is not clear in the official source. If you see the setting in your account, it has.
| Situation | Recommended setting | Why |
|---|---|---|
| Prices, promotions, health/finance/legal wording | Off | Responsibility for accuracy lies with the advertiser |
| Tightly defined brand voice, approval process in place | Off | Unapproved copy breaks the brand voice |
| Small test budget, many variations needed | On per campaign, measured separately | Don't roll it out broadly without seeing its effect |
| The setting doesn't appear in the account at all | Wait | Scope is not clear in the official source |
The one thing to do today: If you have a ChatGPT Ads account, open the settings page in Ads Manager, note the current state of the creative enhancements toggle, and make a deliberate decision by October 29. If you don't have an account, there's nothing to do.
| News | Why people are talking about it | Who it concerns |
|---|---|---|
| Google Ads extended Demand Gen cost adjustments from 3 weeks to 60 days (Search Engine Land) | Adjustments now apply over a longer window; how you look at campaign history changes | People managing Google Ads |
| AI Max was spotted turned on automatically in some Ad Grants accounts (Search Engine Land, October 19) | It's showing up even though account owners didn't turn it on; there are concerns about relevance and click-through rate | Nonprofits, foundations and Google Ads accounts |
| An A/B test toggle was spotted in Performance Max asset groups (Search Engine Land, October 20) | No official announcement, only seen in some accounts | Digital marketing teams running PMax |
| Google removed the "AI Overview" label from some AI Overviews (Search Engine Roundtable, October 20) | Barry Schwartz thinks it may be a bug; Google didn't respond | GEO and content teams |
| Google Business Profile tests verification with government ID (Search Engine Roundtable, October 20) | It's not on Google's help page; which countries it's being tested in is unclear | Local businesses, people managing SEO |
As of Friday, October 30, 2026, they'll be on automatically in new and existing self-serve campaigns. You can turn them off in Ads Manager before or after October 30.
According to the sources, the headline and description, the image size, the language of the ad copy and the destination URL. Versions can be prepared in advance or generated on the spot based on the conversation, and they can be served without your approval.
The sources don't mention Turkey. ChatGPT's ad dashboard has opened in Turkey, but whether creative enhancements have reached accounts in Turkey is not clear in the official source. If the setting appears in your account, assume it applies.
Let's set up your ChatGPT ads account together or review its settings: ChatGPT ad management service. For a quick look at your brand's digital standing: START YOUR FREE ANALYSIS . The previous ChatGPT Ads development: ChatGPT Ads Get a Visual Format and Independent Measurement.
OpenAI, October 5, 2026: an image ad test on ChatGPT's image generation screen (US) and independent measurement partners like AppsFlyer, Adjust and LiveRamp. What changes for advertisers, including in Turkey?
I explain where ChatGPT ads appear and who sees them, where account availability stands for businesses in Turkey today, and how to set up your first campaign step by step. I checked every point against OpenAI's own help pages.
ChatGPT ads opened in 52 countries on September 2, and Turkey isn't on the list. But shopping results are now open only to brands with a connected product feed, and that's where the real action is for businesses in Turkey.
We use cookies to improve the site and measure advertising performance. Analytics and advertising cookies run only if you accept them. Details are in our Privacy Policy and Data Protection Notice.
