Do you have to label AI content on TikTok, Instagram and YouTube?

TikTok's community guidelines ask creators to label AI-made or heavily AI-edited content that shows realistic scenes or people. If you do not, TikTok may remove or restrict the post, or label it itself. A label does not make otherwise harmful or misleading content acceptable, and putting the label on content that was not made with AI is against TikTok's terms.

To label a video, tap Next, then More options, and turn on AI-generated content; for a photo post the switch is in the post settings. Viewers then see that the creator labeled it as AI-generated. TikTok can also add its own label when a post was made with TikTok's AI effects or carries Content Credentials, the open standard for marking how media was made, and a creator cannot remove that one.

TikTok says a labeled post that follows its guidelines is not held back for being labeled.

Meta asks you to disclose photorealistic video and realistic-sounding audio that was made or altered digitally, AI included, and may apply penalties if you do not. Still images do not require it, but Meta labels them anyway when it detects AI through industry-standard signals in the file, such as Content Credentials. Content that was only edited with AI shows that in the post's menu rather than as a label on the post.

To add the label yourself, turn on Add AI label when you share a post or reel; on a story it is under Options. Meta has not said whether a label changes how far a post travels.

YouTube asks you to disclose content that seems realistic: a real person saying or doing something they did not, altered footage of real events or places, realistic scenes that never happened, and AI music that is the main focus. Clearly unrealistic or animated content, beauty filters, colour and lighting fixes, captions, upscaling and help with scripts, titles, thumbnails or ideas are exempt.

You disclose it in YouTube Studio, under the video's AI use setting. The label then shows on the player for photorealistic content, or in the expanded description otherwise. YouTube can add its own label, and creators who keep failing to disclose may have content removed or lose Partner Program access. YouTube says disclosing on its own does not limit a video's audience or its ability to earn money.

The rules are about what the viewer sees and hears, not about who wrote the words. A screen recording of your app needs no label. Neither does a wall of text whose words you wrote with an AI's help over real footage.

AI-made pictures of realistic people or places are what the rules are aimed at. A slideshow of AI-made photos of a realistic person should carry TikTok's label. An AI-made video of a person, or an AI voice that sounds real, should carry the label on all three platforms. When in doubt, label it: every platform's penalties are for leaving it off, and none for adding it to content that really was made with AI.

Apps that publish for you can set the label through each platform's posting API: TikTok's has an AI-generated flag, YouTube's has a flag for altered or synthetic content, and Instagram's added one for AI-generated media. Whether a tool sets them is up to the tool, so ask what yours sets, and remember that labeling correctly stays your responsibility, not the tool's.

Topped starts every post with TikTok's, Instagram's and YouTube's AI-generated label switched on, and you can switch it off before you post one that has nothing AI-made in what the viewer sees. There is a free plan with no card.