Do you have to label AI videos of yourself?
Platform by platform, for videos made with your own face or voice.
You see every video before it posts.
On this page
The short answer
Mostly, yes. TikTok and Meta (Instagram, Facebook, Threads) require a label on realistic AI video or audio of people and make no exception for your own face; LinkedIn requires disclosure when synthetic media shows someone saying what they did not say; YouTube requires it for realistic content but explicitly exempts cloning your own voice; X only bans deceptive media. In the EU, businesses that publish realistic AI video have had to disclose it since 2 August 2026.
Updated
What to actually do
Label realistic AI video of yourself
On TikTok, Instagram, Facebook, Threads, LinkedIn and YouTube (unless only your voice is cloned), and in the EU, label it. It costs nothing and lowers the risk of the post being labelled, restricted or removed for you.
Use the platform’s own toggle
TikTok’s AI-generated content label, Meta’s AI label and YouTube’s “AI use” setting are built for this; a caption works where there is none.
Never borrow someone else’s face or voice
Faking another person is riskier everywhere: TikTok bans AI likenesses of private people without consent, and every platform above acts against deceptive fakes of real people.
Keep the claims true
A label covers how the video was made, not what it says. What your avatar says should be what you would say on camera.
The rules at a glance
| Where | Label required? | Exception for your own face or voice? | How to label |
|---|---|---|---|
| TikTok | Required for realistic AI people | No exception | AI-generated content label, or a clear caption |
| Instagram, Facebook, Threads | Required for photorealistic video and realistic audio | No exception | The AI label when you post (“AI info”) |
| YouTube | Required for realistic content | Own voice clone exempt; own face not addressed | YouTube Studio, “AI use”: Yes |
| Required for synthetic media of a person | Not spelled out | Say it clearly in the post | |
| X | No labelling duty; deceptive media banned | Not addressed | No label required; disclosing is the safe choice |
| EU AI Act, Article 50 | Deployers must disclose deepfakes, since 2 August 2026 | Private use is outside it; businesses are covered | Disclose clearly, at first exposure |
Yes Partly or unclear No
Summarised from each source below, read October 1, 2026. Not legal advice.
The rules, one by one
Quoted or summarised from each platform’s own policy pages and the European Commission, on the date shown. Rules change often; check the source before you rely on one. This is not legal advice.
TikTok
Labels are required for realistic AI content, and unlabelled content can be removed, restricted or labelled by TikTok.
Source: tiktok.com, read October 1, 2026
- Rule
- “We require creators to label AI-generated or significantly edited content that shows realistic-looking scenes or people.”
- Own likeness
- No exception for yourself: likeness includes your face, body and voice, and AI audio that mimics a real person’s voice needs disclosure.
- How to label
- TikTok’s AI-generated content label, or your own clear caption, sticker or watermark.
Instagram, Facebook, Threads
Meta requires disclosure on photorealistic video and realistic audio made or altered with AI, and may apply penalties if you do not.
Source: about.fb.com, read October 1, 2026
- Rule
- “We require people to use this disclosure and label tool when they post organic content with a photorealistic video or realistic-sounding audio that was digitally created or altered.”
- Own likeness
- No exception for yourself in Meta’s wording; a realistic clone of your face or voice falls under it.
- How to label
- The AI label option when you post; Meta shows it as “AI info”.
YouTube
Disclosure is required for realistic content that makes a real person appear to say or do something they did not.
Source: support.google.com, read October 1, 2026
- Rule
- Required when realistic content “makes a real person appear to say or do something they didn’t do”, alters a real event, or shows a realistic scene that did not happen.
- Own likeness
- Explicitly exempt: “cloning one’s own voice to create voice overs or dubs”. YouTube does not address a realistic AI video of your own face.
- How to label
- In YouTube Studio, under “AI use”, choose Yes. Repeated non-disclosure can lead to labels, removal or loss of the Partner Program.
LinkedIn
LinkedIn’s community policies require disclosure of synthetic media that shows a person saying or doing what they did not.
Source: linkedin.com, read October 1, 2026
- Rule
- “Do not share synthetic or manipulated media that depicts a person saying something they did not say or doing something they did not do without clearly disclosing the fake or altered nature.”
- Own likeness
- No exception for yourself; whether your avatar reading your own script counts is not spelled out.
- How to label
- Say it clearly in the post. Files with C2PA Content Credentials show LinkedIn’s credential icon.
X
X has no labelling duty; it bans deceptive media that could cause confusion or harm.
Source: help.x.com, read October 1, 2026
- Rule
- Bans “inauthentic media … that may result in widespread confusion on public issues, impact public safety, or cause serious harm”; non-deceptive sharing is allowed.
- Own likeness
- Not addressed. Media of a real person fabricated with AI is banned when shared as authentic and deceptive.
- How to label
- No platform label required; disclosing in the post is still the safe choice.
EU AI Act, Article 50
Since 2 August 2026, deployers of AI that makes deepfake image, audio or video must disclose that it is artificial.
Source: digital-strategy.ec.europa.eu, read October 1, 2026
- Rule
- “Deployers of an AI system that generates or manipulates image, audio or video content constituting a deep fake, shall disclose that the content has been artificially generated or manipulated.”
- Own likeness
- A person using AI privately is outside it; someone gaining an economic benefit on a regular basis becomes a deployer, so businesses and professional creators are covered.
- How to label
- Disclose clearly, at the latest at first exposure. Artistic, satirical or fictional work only needs a disclosure that does not spoil it.
Read next
Your questions, answered
Short answers to what people ask next.
Yes. TikTok requires a label on AI-generated content showing realistic-looking people, and its definition of likeness includes your own face and voice. Use TikTok’s AI-generated content label or a clear caption.
Yes, for photorealistic video or realistic audio made with AI. Meta asks you to use its AI label when you post, and may apply penalties if you do not.
Not for the voice alone: YouTube explicitly exempts cloning your own voice for voice overs or dubs. For realistic video that shows a real person saying or doing something they did not, it asks you to choose Yes under “AI use”.
Since 2 August 2026, Article 50(4) has required deployers of AI that generates deepfake image, audio or video to disclose that it is artificial. A private person is outside it; a business or professional creator using it regularly is a deployer.
No. It summarises each platform’s and the European Commission’s published rules on 1 October 2026. Check the linked sources, which change often.