ChatGPT Ads creative enhancements are a setting that lets OpenAI rewrite your headlines and descriptions, resize images, change destination URLs and translate ad text, then run those versions without your approval. OpenAI switches it on for new and existing campaigns at the end of October 2026, and you can opt out in Ads Manager per account or per campaign.
This guide is part of our AI advertising hub and follows OpenAI's help center and Ad Tools Terms as of October 2026. The short recommendation: decide per campaign before October 29, 2026. Keep enhancements on where claims are simple and every landing page is tracked, and switch them off where claims are regulated, brand rules are strict or reporting depends on one exact URL. New to the channel? Start with the ChatGPT Ads explainer.
What are ChatGPT Ads creative enhancements?
ChatGPT Ads creative enhancements are automated changes OpenAI makes to an advertiser's ads so they fit the conversation and the user's interests. OpenAI builds them from the existing ad and the landing page content, and the enhanced versions can run without the advertiser reviewing or approving each one, according to OpenAI's help center.
OpenAI's help article on creative enhancements lists three inputs: the conversation context or user interests, your existing ad, and content from your landing page. Variations can be prepared in advance or created in real time based on the conversation. The article's own example turns the headline "Fresh groceries delivered" into "Groceries for quick weeknight dinners."
The contractual basis is older than the toggle. OpenAI's Ad Tools Terms, published August 24, 2026, describe Creative Tools that may generate, modify, optimize, localize or translate ad creatives, and say they can be applied automatically based on your campaign settings. Digiday reported the first version of that language in June 2026.
OpenAI also states that "Ads do not influence ChatGPT's answers." Enhancements change the ad, not the answer next to it.
When do creative enhancements switch on, and which campaigns are affected?
Creative enhancements switch on automatically at the end of October 2026 for new and existing campaigns. OpenAI's sources disagree by one day: the help center says Thursday, October 29, 2026, while the advertiser email names Friday, October 30, 2026. Act before October 29 to be safe, because opt-outs set earlier are kept.
Officially confirmed by OpenAI's help center: "Starting Thursday, October 29, 2026, creative enhancements will be enabled automatically for new and existing campaigns." You can opt out before that date or change the setting at any time, existing account and campaign opt-outs are preserved, and "Campaign-specific settings take priority over your account default."
Reported from OpenAI's email: Search Engine Roundtable published the email on October 20, 2026. It reads: "Starting Friday, October 30, 2026, we'll enable creative enhancements for your new and existing self-serve campaigns." Barry Schwartz added that the toggle already showed as on in his own account settings.
Not documented: whether managed campaigns run by an OpenAI sales team follow the same date. The email says self-serve, the help center does not qualify it. Advertisers with a managed setup should ask their OpenAI contact. Reactions quoted in the same article were skeptical: PPC consultant Melissa Mackey said her team had already paused its ChatGPT Ads test over lead quality.
What can creative enhancements change in your ads?
Creative enhancements can change four parts of a ChatGPT ad: headlines and descriptions, the language of the ad text, image sizing and the destination URL. Each carries a different risk. Copy and language changes affect claims and brand voice, image resizing is mostly technical, and URL changes affect where traffic lands and how it is tracked.
| Enhancement | What OpenAI says it does | Main risk | What to check |
|---|---|---|---|
| Headlines and descriptions | Writes new or updated text that links your offer to the conversation or user interests | Claims, tone and promotions drift from approved copy | Prices, offers and any claim that needs proof |
| Language | May translate your ad text into another language | Translated claims and legal terms no one reviewed | Markets where ad language is regulated or brand specific |
| Images | Resizes images to fit the ad format | Cropped logos, text or product details | Images with text near the edges |
| Destination URLs | May send people to a relevant page on your website | Traffic lands on untagged or unsuitable pages | Tracking, consent and stock on every page |
OpenAI's FAQ is explicit on review: "The experience does not provide a preview of every variation." Results show in Ads Manager reporting at campaign, ad group and ad level. The help center does not say which languages translation covers or which pages URL optimization may pick, so treat both as open.
What are the brand, compliance and tracking risks?
The brand, compliance and tracking risks of ChatGPT Ads creative enhancements come from one fact: versions run without per-version approval, while the advertiser stays responsible for them. Rewritten claims can break ad policy or local law, translated text can reach markets nobody reviewed, and a changed destination URL can drop tracking parameters or land on a page without the pixel.
Responsibility. Section 3 of the Ad Tools Terms makes you responsible for the accuracy and substantiation of claims in generated creatives applied to your campaign. OpenAI does not guarantee they are accurate or legally compliant.
Policy. OpenAI's ad policies say "Ads must be consistent end-to-end" and ban unfounded claims about pricing, outcomes or comparisons. Health and financial ads are restricted, so a rewritten claim there carries real policy risk.
Tracking. OpenAI's help article on creating ads says you add your own UTM parameters to each destination URL. The measurement pixel docs say the pixel reads the oppref click ID from the landing page URL and stores it for 30 days. OpenAI does not document whether an optimized URL keeps your UTM tags. If it does not, paid visits can show up in GA4 as another channel, a risk covered in our guide to ChatGPT Ads conversion tracking.
How do you opt out of creative enhancements in Ads Manager?
You opt out of ChatGPT Ads creative enhancements in Ads Manager, either as the account default or per campaign. OpenAI's help center points to a "Manage settings" control and says campaign settings override the account default. Opt-outs set before the switch date are preserved, so the safest moment to decide is before October 29, 2026.
- List your campaigns by risk. Mark every campaign with regulated claims, strict brand rules, single-language markets or one exact tracked URL.
- Open the account settings. Search Engine Roundtable shows the creative enhancements toggle on the ad account settings page in Ads Manager.
- Set the account default. Choose off if most campaigns are on your risk list, on if only a few are.
- Override per campaign. Change the setting on each campaign that differs from the default. OpenAI does not publish the exact menu path, so check the campaign settings in your own account.
- Record the decision. Note per campaign whether enhancements are on and why.
- Check after the switch date. On October 30, open the settings again and confirm nothing flipped back on.
The Ad Tools Terms state that "You can manage or disable Creative Tools through applicable Ads Manager controls." No API or bulk sheet option for this setting is documented as of October 2026.
Should you keep ChatGPT Ads creative enhancements on or switch them off?
Keep ChatGPT Ads creative enhancements on for campaigns with simple, unregulated claims, a single language market and every reachable page tagged and tracked. Switch them off for regulated categories, legally reviewed copy, deliberate language choices or reporting tied to one exact landing page. When in doubt, test one campaign rather than the whole account.
| Campaign situation | Decision | Why |
|---|---|---|
| Simple offer, no regulated claims, pixel on every page | Keep on, review weekly | Low claim risk and tracking survives a URL change |
| Health, financial or other regulated claims | Switch off | You carry the liability for rewritten claims |
| Copy approved by legal or a brand team | Switch off | No preview of every variation exists |
| Ads targeted at one language on purpose | Switch off | Translation may produce text nobody reviewed |
| Reporting depends on one exact tagged URL | Switch off | URL handling of UTM tags is undocumented |
| Testing whether the channel works at all | Keep off in the test cell, on in a second cell | Gives a cleaner read than a before and after view |
How do you test creative enhancements without risking the whole account?
You test ChatGPT Ads creative enhancements by turning them on in one low-risk campaign and comparing it with a matching campaign that keeps them off over the same weeks. OpenAI itself warns that changes over time "do not, on their own, establish the effect of creative enhancements," so a before and after view is not enough.
Hypothetical worked example: a retailer runs ChatGPT Ads in Belgium with separate Dutch and French campaigns and tracks each campaign by a tagged category URL. Under the per-campaign decision rule for ChatGPT Ads creative enhancements (switch off for single-language markets and for reporting tied to one exact URL), the account default goes off, because both translation and URL changes collide with how the account is built. One low-risk campaign gets enhancements on for four weeks, with its pages tagged and the pixel checked, and is compared with a matching campaign that stays off.
Brian Quinn of AppsFlyer told Digiday in June 2026 that "the real test will be campaign performance." That applies here too: OpenAI publishes no performance figures for enhancements. The same trade-off exists on other platforms, as our guides to TikTok Smart Creative and AI ads show.
Common mistakes with ChatGPT Ads creative enhancements
Common mistakes with ChatGPT Ads creative enhancements come from treating an account-wide default as a decision and from assuming the platform will show you every version. Most are fixed in minutes in Ads Manager before October 29, 2026, and each is cheaper to prevent than to explain to a legal team or client afterward.
- Waiting for October 30. The help center names October 29. Set opt-outs earlier; they are kept.
- Only changing the account default. Campaign settings override it. Check every campaign that already has its own setting.
- Expecting a preview. OpenAI provides no preview of every variation. If approval matters, the setting has to be off.
- Tagging only the main landing page. URL optimization can send traffic elsewhere on your site. Tag and track every page an ad could reach.
- Reading a before and after change as proof. OpenAI says changes over time do not on their own show the effect. Use a split between campaigns instead.
- Forgetting product feed campaigns. Check whether the setting appears there too, since ChatGPT Ads product feed campaigns build ads from templates and your catalog.
