AI WORLD HQHow to label AI-made videos when you post them on YouTube, TikTok and Facebook
When a video made or changed with AI has to be marked as AI, and where each platform's own help pages say to turn the label on before you post.
This guide was written by AI. It passed automated fact and quality checks, and our editor spot-checks published guides every day.
See AI World HQ first on GoogleAdd us as a preferred source and our stories show up higher in Google’s Top Stories.Add on GoogleMore and more videos are made or edited with AI: a cloned voice-over, a generated scene, a face swapped into a clip. Viewers deserve to know when something that looks real isn't. YouTube, TikTok and Meta each have their own rules and their own label. This guide describes each one separately, based on that company's own help pages and announcements, so you can follow the rules wherever you post.
💡 Tip: tap a step’s number when you finish it — a green tick appears and your browser remembers how far you got.
- The finished video, and a clear idea of which parts were made or changed with AI.
- Your account on the platform where you'll post.
- About 5 minutes per platform.
Decide whether your video needs a label
Each company sets its own rules. In general, realistic AI content needs a label, and small edits usually don't. Here is what each one says.
YouTube says to disclose content that:
- makes a real person appear to say or do something they didn't,
- alters footage of a real event or place, or
- shows a realistic-looking scene that didn't actually happen.
YouTube also asks you to disclose AI music when it's the main focus of the video. It says you don't need to disclose clearly unrealistic or animated content, beauty filters, color or lighting fixes, captions, or using AI only to help write your script, title, or thumbnail.
TikTok asks you to label content that was completely generated or significantly edited by AI.
Meta (Facebook) requires its label for photorealistic video or realistic-sounding audio that was digitally created or altered.
you can say in one sentence what in your video is AI-made.
Facebook: use Meta's AI label before you post
Meta says people must use its AI disclosure and label tool when they post a photorealistic video or realistic-sounding audio that was digitally created or altered, and that it may apply penalties if they don't. Meta shows these posts with an AI info label, and it may add the label itself when it detects industry-standard signals that content was made with AI.
Meta doesn't publish step-by-step screens that we could verify, so here's the safe approach: before you post, look for the AI label option in your post or reel settings and turn it on.
your published post shows the AI info label.
**** If you can't find the option, check Meta's Help Center for the current steps before posting.
- Thinking "it's only a little AI" means no label: A realistic cloned voice or a swapped face counts, even if the rest is real footage.
- The fix: Use Step 1. If the AI part could make viewers believe something happened that didn't, label it.
- Labeling on one platform and forgetting the others: Each platform has its own switch.
- The fix: When you post the same video in several places, turn the label on in each one.
- Hiding the AI part in the caption only: A caption alone may not count as the platform's label.
- The fix: Use the official switch. You can also mention it in the caption.
- Using someone else's face or voice without permission: A label doesn't make that okay.
- The fix: Only use people's likeness or voice with their clear consent.
Pick one video you plan to post this week. Write down in one sentence what in it was made with AI, check it against Step 1, and if it needs a label, turn on the switch from Step 2, 3, or 4 when you upload it.
🔧 If you get stuck
Can I check whether my video was made with AI before I decide on a label?
Why: Gemini can currently only recognize content created by Google AI tools. Other companies have started to adopt SynthID watermarks, but Gemini can only detect Google's.
What to do: In Gemini: you need a personal account or qualified Workspace account, and you must be signed in to Gemini Apps before you verify content. On your computer, go to gemini.google.com, click Add files, and click Upload Files to add a file from your device. Then ask a verification question such as "Was this video created or edited by Google AI?"
Source: Verify AI-generated images, videos, and audio - Computer
Does Gemini still detect AI if I resized or compressed the video?
Why: The digital watermark will usually still exist even if the image, video, or audio is re-scaled, re-colored, compressed or altered in other ways. However, there's still a chance that after many alterations the watermark won't be detected.
What to do: In Gemini: a SynthID watermark will usually still exist after the video is re-scaled or compressed. After many alterations, the watermark may not be detected.
Source: Verify AI-generated images, videos, and audio - Computer
❓ Quick questions
How long does this take?
About 6 minutes — the guide has 4 steps, and you can tick each one off as you go.
Do I need to prepare anything?
- The finished video, and a clear idea of which parts were made or changed with AI.
- Your account on the platform where you'll post.
- About 5 minutes per platform.
What mistakes should I avoid?
- Thinking "it's only a little AI" means no label: A realistic cloned voice or a swapped face counts, even if the rest is real footage.
- The fix: Use Step 1. If the AI part could make viewers believe something happened that didn't, label it.
- Labeling on one platform and forgetting the others: Each platform has its own switch.
- The fix: When you post the same video in several places, turn the label on in each one.
- Hiding the AI part in the caption only: A caption alone may not count as the platform's label.
- The fix: Use the official switch. You can also mention it in the caption.
- Using someone else's face or voice without permission: A label doesn't make that okay.
- The fix: Only use people's likeness or voice with their clear consent.
Keep reading
📬 The week’s AI, in your inbox
One friendly email every Sunday — the 5 stories that mattered, in plain English. No spam, unsubscribe anytime.
✦ Generated by AI in AI World HQ's automated newsroom, written in plain language. Checked by automated fact and quality gates, and our editor spot-checks published guides every day. Spot a mistake? Use the buttons above.
📘 Prefer it in one file? Photos and Images: The AI Pack (20 guides, free printable PDF)
← Back to all stories