YouTube Altered or Synthetic Content: Disclosure Guide

Direct answer

YouTube asks creators to disclose content that is realistic and generated or meaningfully altered with AI. Its documented Studio workflow uses the AI use setting under Attributes, but the correct answer depends on what viewers would believe happened. YouTube explicitly excludes some production assistance and cloning your own voice for a voice-over or dub from the examples that need disclosure. Assess EU law and other platform rules separately.

Practical decisions

Synthetic narrator, voice clone, realistic scene or production help?

Each case describes a specific final asset. The platform decision is separate from legal duties, rights and client policy.

Generic synthetic narrator over real footage

Asset: A non-identifiable generated voice reads a product script over real filmed scenes.

Platform decision: A synthetic voice is not automatically a clone of a real person. Decide whether the finished video meets YouTube's realistic, meaningful-alteration test; do not apply a blanket rule based only on the tool.

Action: Document the narrator's identity status and review the AI use setting against the final video. Use component-specific copy if disclosure is otherwise required.

Evidence to keep: Keep final audio and video, decision rationale and the published setting/label state.

Another person's voice is cloned for an endorsement

Asset: The video makes an identifiable real person appear to recommend the product without recording those words.

Platform decision: YouTube lists altering a person's voice to make them say something they did not say among disclosure cases. A label does not cure missing consent or a deceptive endorsement.

Action: Obtain permission and review the claim before publication; if allowed, select Yes for AI use and disclose the synthetic speech.

Evidence to keep: Retain consent, final cut, reviewer approval and visible post-publication label.

A realistic event scene is invented

Asset: AI creates footage of a real location or event that did not occur as shown.

Platform decision: This is a meaningful realistic alteration under YouTube's examples and needs AI use disclosure.

Action: Select Yes under AI use and add context that distinguishes the synthetic scene from recorded footage.

Evidence to keep: Keep source shots, generated-scene record, selected setting and delivered-video capture.

AI only helps prepare the production

Asset: A team drafts a script or thumbnail with AI, repairs recorded audio or clones its own voice for a dub.

Platform decision: YouTube lists these among examples that do not require disclosure merely for that assistance. A later realistic generated scene or third-party voice clone changes the facts.

Action: Record the actual final-media use; answer the AI use question for the published video, not the production tools in isolation.

Evidence to keep: Keep a concise AI-use manifest and the final upload review.

Do you need disclosure for your asset?

Check the final media, AI use, audience and destination separately. This is not an inspection of the file's metadata.

Check disclosure

First-party policy text was checked on 23 September 2026. These are documented workflows, not a live test of each account, region, post format, ad placement or scheduler interface. Verify the actual publishing UI and served content. See the source register.

Applies when

Realistic altered or synthetic content that could mislead viewers.
Trigger scope: Realistic altered or synthetic content.
Video: Select Yes or No under Attributes → AI use in the documented Studio upload flow.
Synthetic voice: Assess whether the final speech is a realistic, meaningful alteration; own-voice dubbing is a stated exception.
Realistic scenes: Disclose realistic generated or meaningfully altered events and places.

Doesn't apply when

Production assistance: Script, title, thumbnail and repair assistance alone are listed as not requiring disclosure.

Worked decision

Example

Generic synthetic narrator over real footage

Assess the final asset and delivered label

A synthetic voice is not automatically a clone of a real person. Decide whether the finished video meets YouTube's realistic, meaningful-alteration test; do not apply a blanket rule based only on the tool.

Next steps

Required action

In YouTube Studio, review Attributes → AI use in the documented upload flow.
Select Yes when the final video is realistic and generated or meaningfully altered with AI; otherwise select No.
Keep component-specific context for viewers when a platform label alone would be unclear.
Publish and verify whether the label appears in the description or player; YouTube controls placement.

Publishing layer

Platform differences

EU legal layer

EU AI Act Article 50(4) requires deployers to disclose realistic AI-generated or manipulated image, audio or video resembling real persons, objects, places or events. Article 50 obligations apply from 2 August 2026.

Scheduler and API

The tracked help page documents Studio's AI use setting; equivalent behavior for every API or scheduler path is not established here. Verify the final video in Studio.

Known limitation

YouTube decides label prominence based on sensitivity (news, elections, health, finance); creators cannot force the prominent variant.

Evidence

Source

Disclosing altered or synthetic content — YouTube HelpYouTube · effective from 2024-03-18 · last checked 2026-09-23
How we're helping creators disclose altered or synthetic contentYouTube · effective from 2024-03-18 · last checked 2026-09-10
Regulation (EU) 2024/1689 (Artificial Intelligence Act)European Union · effective from 2024-08-01 · last checked 2026-09-12

Freshness and accountability

Last verified

Last reviewed 2026-09-23 · Current

Run this exact scenario

A representative scenario and YouTube are preselected. Confirm the final media, AI operation and audience.

Open prefilled checker

Native AI-label capability

Native label	full AI use disclosure
Applies to	Realistic altered or synthetic content that could mislead viewers.
Surface	YouTube video uploads
Trigger scope	Realistic altered or synthetic content
Permanence	Platform-controlled
Reversible after publication	not verified
Rollback path	No verified path recorded
Rollback confidence	unknown
Post-publish edit	unknown
Label state verifiable	not verified
Legal threshold	independent Signals realistic altered or synthetic content under YouTube policy.
API / scheduler	partial The tracked help page documents Studio's AI use setting; equivalent behavior for every API or scheduler path is not established here. Verify the final video in Studio.

Content and media types

Type	Native handling	Manual action
Video	native	Select Yes or No under Attributes → AI use in the documented Studio upload flow.
Synthetic voice	native	Assess whether the final speech is a realistic, meaningful alteration; own-voice dubbing is a stated exception.
Realistic scenes	native	Disclose realistic generated or meaningfully altered events and places.
Production assistance	manual	Script, title, thumbnail and repair assistance alone are listed as not requiring disclosure.

Platform policy can change independently of EU law. Disclara records both layers and never treats a native label as a legal-compliance verdict.