Meta ad rejected · AI-generated and digitally altered content
Why was my AI-generated Meta ad rejected, and do I need to disclose AI?
Ads about social issues, elections or politics must disclose realistic images, video or audio made or edited with generative AI, and any ad can be rejected if realistic altered media could seriously mislead people on a matter of public importance, or if it uses a public figure to bait people into a scam. Disclose where required, remove misleading or rule-breaking AI media, then resubmit or request another review in Meta Business Support Home.
What usually triggers it
- A social issue, electoral or political ad with realistic AI-made or AI-edited media that was not disclosed, such as a real person saying something they did not say, a realistic person who does not exist, or an event that did not happen.
- A photorealistic image or video, or realistic audio, that was digitally created or altered and could seriously mislead people on a matter of public importance. Meta can reject this in any ad.
- Falsely using a public figure's name, likeness or imagery, for example in an AI-made video, with misleading tactics to bait people into a scam. Meta calls this celebrity-bait.“As recommended by [famous presenter]: claim yours before it's gone.”
- AI images that break other rules, such as suggestive or nude imagery, or a realistic-looking person pointing a weapon at the viewer.
- Falsely claiming to represent, or speak in the voice of, an established business or brand, with misleading tactics to bait people into a scam. Meta calls this brand-bait.
How to fix the ad
- Work out whether the ad is about a social issue, an election or politics. If it is, complete Meta's authorisation and add a “Paid for by” disclaimer where required, and use Meta's AI disclosure when the ad includes realistic AI-made or AI-edited media of the kinds Meta lists.
- Remove any realistic AI or edited media that shows a real person saying or doing something they did not, or a real-looking event that did not happen, if it could mislead people.
- Never use a public figure's name, face or voice to imply an endorsement you do not have. Use your own customers or staff, filmed with their permission.
- Check AI images against the same rules as photos: no nudity, 18+ targeting for suggestive or body-focused images, no gore, and no realistic weapons pointed at the viewer.
- An “AI Info” label is not a rejection. Meta's political ads policy says that from 1 June 2026 it also detects ad media made or edited with third-party generative AI tools and shows the label in “About this ad”, with no action needed from the advertiser.
- Run the copy and images through our free checker at /check before upload. It estimates risk; it is not Meta's decision.
Risky wording, and safer wording
Illustrative examples to show the difference. Always check the full ad, image and landing page too.
| Risky | Safer |
|---|---|
| An AI video of a famous TV presenter recommending your skincare. | A real customer, filmed with permission, talking about how they use your skincare. |
| A realistic AI image of a candidate at an event that never happened, with no disclosure. | A genuine photo from the candidate's public event, run with authorisation and a “Paid for by” disclaimer. |
| A realistic AI image of flood damage presented as this week's news, to sell home insurance. | Photos of your own team at work, with plain copy about what the policy covers. |
| An AI model posed suggestively in your swimwear. | An AI model wearing the swimwear in a natural, non-suggestive pose. |
If your ad is already rejected
- Open the rejection email or Meta Business Support Home and read the reason. Check whether it names missing AI disclosure, a missing disclaimer, misinformation, or another policy such as adult or violent content.
- Edit the ad in Meta Ads Manager to add the disclosure or replace the media. An edited ad is automatically sent for review again. Avoid deleting it if you might want a review, because Meta says a deleted ad may lose the option to request one.
- If you believe the ad did not break the policy, request another review in Business Support Home: select the account, select the ad, click Request review, then Submit. Meta aims to decide within 48 hours, and if the ad is rejected again you cannot request another review for it.
- Do not re-upload the same media in new ads without the disclosure. Meta says repeated failure to disclose may result in penalties, and editing or deleting a rejected ad does not remove the violation.
- Keep records: the tools used to make each asset, any permissions, a screenshot of the rejected ad, the reason Meta gave and what you changed.
Check your next ad before you upload it
Paste the copy and add the image or video. The free checker scores the rejection risk, points to the exact words at fault and the Meta policy behind each one, and suggests a safer version. It estimates risk; it isn't a decision from Meta.
Likely to be rejected. Fix the flagged lines before you upload.
This supplement ad combines a personal-attribute callout, a specific weight-loss result, guaranteed results, clickbait and a URL shortener, so it is very likely to be rejected.
Are you struggling with belly fat? Our gummies helped Maria lose 12 kg in 3 weeks!!! See the before and after pics. Doctors hate this one trick.
GUARANTEED RESULTS
https://bit.ly/slim-offer
6 issues found · tap a highlight to jump to it
“Are you struggling with belly fat?” · primary text
The ad asserts a personal characteristic about the viewer by addressing them directly about a body condition.
Fix: Our gummies are designed to support a healthy metabolism.
“helped Maria lose 12 kg in 3 weeks!!!” · primary text
A specific weight-loss amount within a short timeframe is an exaggerated, unrealistic result, and supplement ads must target 18+.
Fix: Remove the specific result; use a general statement such as 'Part of a balanced routine for healthy weight management.'
“GUARANTEED RESULTS” · headline
Guaranteed results are a prohibited claim for health and weight-loss products.
Fix: Use 'Start your wellness routine today' or remove the claim entirely.
“Doctors hate this one trick.” · primary text
This is classic clickbait, and the triple exclamation marks add to the sensational tone.
Fix: Replace with 'Learn how our gummies fit into a healthy routine.'
“See the before and after pics.” · primary text
Before-and-after comparisons for weight loss are high risk and draw rejection.
Fix: Remove the before-and-after reference and focus on the product's ingredients.
“https://bit.ly/slim-offer” · landing page
URL shorteners hide the true destination and are high risk, and can lead to account-level enforcement.
Fix: Use the full, direct landing page URL on the brand's own domain.
Safer version
Are you struggling with belly fat? Our gummies helped Maria lose 12 kg in 3 weeks!!! See the before and after pics. Doctors hate this one trick.
GUARANTEED RESULTS
Meet our gummies, designed to support a healthy metabolism. Easy to add to your day and made to fit into a balanced routine for healthy weight management. Learn how they fit into a healthy routine.
Start your wellness routine today
Also change: Link: replace the bit.ly shortener with the full, direct URL on your own brand domain, and make sure the landing page has no before-and-after photos, guaranteed-result claims or specific weight-loss numbers. Creative: remove any before-and-after images and any overlay text about kilos lost. Show the product or lifestyle shots instead, with overlay text like "Support your healthy routine". Set ad targeting to 18+.
Questions
Do I have to disclose AI in every Meta ad?
No. The disclosure requirement in Meta's ad policies covers ads about social issues, elections or politics that contain realistic AI-made or AI-edited images, video or audio of the kinds it lists. Any ad can still be rejected if realistic altered media could seriously mislead people on a matter of public importance.
What is the AI Info label on ads?
Meta's policy on social issue, electoral and political ads says that from 1 June 2026 it also uses automated detection to spot ad media made or edited with third-party generative AI tools. When it does, an “AI Info” label appears in “About this ad”, reached from the three-dot menu, and no advertiser action is needed.
What happens if I don't disclose AI in a political ad?
Meta says it will reject the ad, and repeated failure to disclose may result in penalties against the advertiser.
Can I use AI to show a celebrity using my product?
Not without a real endorsement. Meta prohibits celebrity-bait, which means falsely using a public figure's name, likeness or imagery with misleading tactics to bait people into a scam, and it says it will enforce against more celebrity-bait in ads over the coming months.
Are AI images treated differently from photos under other policies?
Treat them the same. Meta's nudity rules refer to photorealistic or digital imagery, and its violent content rules cover fictional people in images that look like real photos, so AI creative gets no exception.
Meta's official pages
- Ads about Social Issues, Elections or Politics (Meta Transparency Center)
- Misinformation, Community Standards (Meta Transparency Center)
- Prohibited Commercial Practices (Meta Transparency Center)
- Our Prohibited Commercial Practices policy (Meta Transparency Center)
- How to troubleshoot a rejected ad (Meta Business Help Center)
Ad Account Shield is an independent tool. It is not affiliated with, endorsed by or sponsored by Meta.
Related guides
- MisinformationFact-checked claims, harmful health or vaccine claims, false voting details or misleading altered media.
- Special Ad CategoriesFinancial products and services, employment, housing, and social issues or politics: when you must declare the category and how.
- Misleading claimsPricing, endorsements or promises in the ad could mislead people for money.