Do Meta, YouTube and Google label, reject or penalise AI ads?

What does Meta do with AI ads on Instagram and Facebook?

Meta's help page describes labelling, not a ban. Ads fully created or materially edited with Meta's generative AI creative features carry AI info, usually in the three-dot menu at the top right of the ad. The label moves next to Sponsored when the ad has an AI-generated photorealistic human. Meta also says the feature is being introduced gradually, so not every ad that used its tools will show it⁴.

Rejection is explicit for one group of ads. If a social issue, election or political ad has a photorealistic person who does not exist, a real person shown saying or doing something they did not, or a realistic event that did not happen, made with third-party generative AI, the advertiser must disclose it, and Meta rejects the ad if they do not¹.

Third-party tools are the grey area. Meta's help page lists ads made with third-party creation or AI tools among those that will not carry AI info under that mechanism⁴. Meta's update of 1 June 2026 then says it is adding more information to ads it detects were created with third-party AI tools⁶, and the political ad standard says detected AI media gets an AI Info entry under About this ad, with the note that the feature may not be available in every region¹. Assume detection can happen on any ad.

Meta's misinformation standard adds a general rule. Content that is photorealistic and digitally created or altered, and that creates a particularly high risk of materially deceiving the public on a matter of public importance, may be labelled, or rejected if it is submitted as an advertisement⁷. For organic posts, Meta requires disclosure of photorealistic AI video or realistic-sounding audio and says it may apply penalties if people fail to disclose⁷.

Several blogs quote a share of ads rejected for undisclosed AI. We could not find any such figure on a Meta page, so we do not repeat it.

What does YouTube require?

YouTube's disclosure page covers videos that creators upload. Creators must disclose realistic content that makes a real person appear to say or do something they did not, alters footage of a real event or place, or generates a realistic scene that did not happen. Unrealistic content, such as a person riding a unicorn, and minor edits like beauty filters, colour correction, captions, upscaling, scripts and idea generation need no disclosure².

The page says disclosure will not limit audience or eligibility to earn money, and that creators who consistently fail to disclose may see manual labelling or penalties, including removal or suspension from the YouTube Partner Program². It does not discuss paid ads. Ads you buy through Google Ads fall under the Google Ads policies below.

What do Google Ads policies say about AI?

Google's July 2026 update permits advertisers to add text or visual labels directly inside image and video creatives that were generated or modified with AI. Google may also label some assets made with its own AI tools. An AI label setting is rolling out gradually across Google Ads, Display & Video 360, Campaign Manager 360, Merchant Center and Ads Editor. Google adds that using the setting does not guarantee compliance with specific regulations, and it names the European Union, India and New York as places whose AI rules require disclosure on some ads³.

Election advertisers must tick the Altered or synthetic content checkbox for ads with synthetic or digitally altered content. For some formats, such as mobile feeds, mobile Shorts and in-stream, Google generates the in-ad disclosure from that checkbox; for other formats the advertiser must provide a prominent disclosure⁸.

Public figures are the sharpest edge. Google lists business founders and CEOs among public figures, asks advertisers to hold documented consent before using their name, image, video or voice to suggest an endorsement, and gives AI-generated videos of popular YouTubers as an example of a misleading use⁵.

So will my AI ad be labelled, rejected or penalised?

Across the pages we opened, none bans AI video in ads as a category. The risks sit in what the ad shows and claims.

• Labelled: Meta ads made or materially edited with its tools, and detected AI media; some Google assets made with Google's AI tools; YouTube videos where the creator discloses.
• Rejected: Meta social issue, election and political ads that hide realistic third-party AI media, and high-risk deceptive manipulated media submitted as an ad.
• Penalised: YouTube creators who consistently fail to disclose, and Google Ads accounts that falsely imply a public figure's endorsement²⁵.

What should an advertiser do?

Policy pages change, and Meta's own help page says its rollout is gradual, so re-read the relevant page on the day you launch.

• Use each platform's AI label setting where it exists, and add your own on-screen label where a local rule requires one.
• Never show a real person saying or endorsing something they did not, and hold written consent for anyone you do show.
• Keep a record per ad of the tool, plan and date used.
• Check the local rule as well as the platform rule; see our answer on AI ad labelling in India.