Social media

YouTube altered or synthetic content label: when an AI architecture film needs it and how to set it

YouTube's altered or synthetic content label for AI architecture films: when to answer Yes under AI use, what changes for Shorts, and the cost of skipping it.

9 min read

In this article

You're uploading a 30-second AI film of a house your studio designed, and YouTube Studio asks a yes-or-no question about AI use. Answer No on a film that looks like real footage and you risk a label added for you, or worse; answer Yes on everything and your conventional renders get labeled too. YouTube's rule turns on one word, realistic, and architecture films sit right on that line. This guide covers what YouTube counts as altered or synthetic content, which architecture films need it, how to set it in Studio, what's different for Shorts, when YouTube labels a video itself, and what happens if you don't disclose.

Key takeaways

  • YouTube requires disclosure when realistic AI content "generates a realistic scene that didn't actually occur" or "alters footage of a real event or place," which covers a photoreal AI film of an unbuilt project and an AI-animated film of a real one.
  • Clearly unrealistic content, such as "fantastical scenes" and "animated videos," and minor edits like "color adjustment or lighting filters" don't need it; neither does AI used for thumbnails or scripts.
  • You disclose during upload in YouTube Studio: in the Attributes section, under "AI use," select Yes; the steps are the same on a computer and in the mobile app.
  • YouTube labels content made with its own AI tools or carrying C2PA metadata automatically, and those labels "can not be adjusted"; creators who consistently skip disclosure "may be subject to manual application of a label, or penalties."
  • YouTube says disclosure "alone won't limit a video's audience or impact its eligibility to earn money," so there's no reach reason to leave it off.

Does an AI architecture film need YouTube's altered or synthetic content label?

Yes, if it looks real. YouTube requires creators to disclose AI-generated or altered content that appears realistic, including content that "generates a realistic scene that didn't actually occur" or "alters footage of a real event or place" (YouTube Help, as of October 2026). A photoreal AI video of a building that doesn't exist yet, or AI motion and people added to photos of a finished project, both qualify, so select Yes under "AI use" when you upload. A stylized or obviously animated concept doesn't need it.

The rest of this guide shows how to make that call for each kind of film your studio might post, and where the setting lives.

What YouTube counts as altered or synthetic content

YouTube's help page lists four situations that need disclosure when the result looks realistic: content that "makes a real person appear to say or do something they didn't do," "alters footage of a real event or place," "generates a realistic scene that didn't actually occur," or "creates music that's the main focus of the video." It lists as not needing disclosure unrealistic content such as "fantastical scenes" and "animated videos," minor edits including "beauty filters," "color adjustment or lighting filters" and "special effects filters," and production help like scripts, thumbnails and infographics. All as of October 2026.

Applied to the films an architecture or interior studio actually posts:

Film Realistic? Disclose under "AI use"? Why
Photoreal AI film of an unbuilt house, made from renderings Yes Yes A realistic scene that didn't occur
AI-animated film of a finished project from photos, with generated motion and people Yes Yes Alters what a real place looked like; adds people who weren't there
AI sky replacement or removed scaffolding in drone footage of a real building Yes Yes, unless trivial Alters footage of a real place
Conventional Enscape, Lumion or V-Ray walkthrough Yes, but not AI-generated No Not generative AI content; still caption unbuilt work as a concept
Clay-style or watercolor AI concept animation No No Clearly unrealistic
Real footage with an AI-generated thumbnail Footage is real No Thumbnails are listed as production assistance
Real footage with color grading Footage is real No Lighting and color adjustment are minor edits

The two middle cases cause the most hesitation. A film of a real, finished building still needs disclosure when AI generated the camera moves and the people, because the result shows the place in a way it was never filmed. If you're unsure, the test YouTube's examples point to is simple: could a viewer reasonably take this for footage shot on site? If yes, select Yes.

Music is a separate trigger. If AI generated a song that's "the main focus of the video," that alone needs disclosure; a quiet background track under a project film is a different case. Music for architecture videos covers licensing.

How to set the AI use disclosure in YouTube Studio

YouTube's help page gives one set of steps for a computer and the mobile app, as of October 2026:

  1. Go to YouTube Studio and start an upload.
  2. Follow the upload steps to the details.
  3. In the Attributes section, under "AI use," select Yes if your content meets the disclosure requirements, or No if not.
  4. Continue with the rest of the upload.

When you select Yes, YouTube says "the content will be labeled as AI generated or altered for viewers." Where the label shows depends on the content: "a label in the video player may also appear" for photorealistic content, while labels for some non-photorealistic AI content "may appear in the expanded description."

Two habits make this reliable for a studio. Put the AI-use decision on your upload checklist next to the title and thumbnail, so it's never left at the default. And decide by film type, not per upload: if every film from a given tool is photoreal AI, the answer for that tool is always Yes. YouTube Shorts for architects has the rest of the upload checklist, including titles that get found in search.

Shorts: what's different

The disclosure is the same for a Short: the steps on the help page apply to uploads from the mobile app as well as from Studio on a computer. What changes is how much of the label viewers see. In the Shorts feed there's no description panel on screen, so a label in the player is the one viewers notice, and your own disclosure line has to be in the title or on the frame, not buried in the description.

Shorts made with YouTube's own AI tools are labeled automatically, and the help page says content made with those tools "can not be adjusted." If you generate a background or clip inside YouTube and combine it with your project film, expect the label whatever you select.

A project series posted as Shorts, such as one film per stage of a build, needs a consistent answer across episodes; Shorts series for project films shows how to structure one. If the series mixes real site footage with AI scenes, disclose on the episodes that include the AI scenes.

When YouTube labels a video for you

YouTube doesn't rely only on your answer. Its help page says it applies labels automatically to content made with its own AI tools, content containing C2PA metadata, and content it detects as AI-generated. "Creators will have the ability to change AI disclosure in most cases by selecting No," but "content made with YouTube's AI tools, content containing C2PA metadata, or content labeled after manual review can not be adjusted."

So the file you upload can speak before you do. Some AI video tools embed Content Credentials (C2PA) in their exports and some don't; check yours once, in a verification tool, before you rely on either outcome. What a content credential is and who reads it explains how to check an MP4 and what the result means. A label you can't remove next to a description that doesn't mention AI looks worse than a disclosure you made yourself.

What happens if you don't disclose

YouTube's help page names the consequences: "Creators who consistently choose not to disclose this information may be subject to manual application of a label, or penalties from YouTube," and non-disclosure can lead to "removal of content or suspension from the YouTube Partner Program." Partner Program suspension matters less to a studio that isn't monetizing, but removal of a project film and a channel with a record of non-disclosure both matter to anyone using YouTube for credibility.

You select Film is photoreal AI Likely outcome
Yes Yes Label shown; reach and monetization not limited by the disclosure itself
No Yes, and the file carries C2PA metadata Automatic label you can't adjust
No, repeatedly Yes Possible manual labels, penalties, removal or Partner Program suspension
No No (real footage, minor edits) No label needed

The trade is lopsided. YouTube states that "disclosing AI content alone won't limit a video's audience or impact its eligibility to earn money," so the only cost of answering Yes is a label that tells viewers the truth.

Write the description so it agrees with the label

YouTube's label says AI was involved; it doesn't say what's real. For architecture, the useful line is about the project's status. Write the first line of the description, and for a Short the title or an on-screen note, so it agrees with the label:

  • Unbuilt project: "AI-generated visualization of a proposed house. Not built."
  • Finished project, AI-animated: "AI-generated video made from photographs of the completed project. Motion and people are generated."
  • Mixed footage: "Site footage with AI-generated scenes of the finished design."

If you make the film with Arch2Video, select Yes: the tool turns six project images into a photoreal 1080 × 1920 film of about 30 seconds, AI motion interprets your images, and every film includes a small cast of original, fictional adults in at least one scene, so people in your photos are replaced, not animated. Present unbuilt work as concept work. If the film also reaches viewers in the EU, the AI Act's own labeling duty applies on top of YouTube's; see EU AI Act Article 50 for design studios.

Frequently asked questions

Will labeling my video as altered or synthetic hurt its views?

YouTube says no: "Disclosing AI content alone won't limit a video's audience or impact its eligibility to earn money" (YouTube Help, as of October 2026). The label changes what viewers are told, not how the video is distributed. Leaving it off on a photoreal AI film carries the real risk: an automatic label you can't adjust, or penalties for repeated non-disclosure.

Do I need to disclose AI renders of a building that hasn't been built?

If the result is a realistic video, yes. YouTube requires disclosure for content that "generates a realistic scene that didn't actually occur," and a photoreal film of an unbuilt house is exactly that. Conventional renders from Enscape, Lumion or V-Ray aren't generative AI content, so they don't trigger the setting, though an "unbuilt concept" line in the description is still the honest thing to write.

Can I remove the AI label YouTube added to my Short?

Sometimes. YouTube says creators can change the AI disclosure "in most cases by selecting No," but content made with YouTube's AI tools, content containing C2PA metadata, and content labeled after manual review "can not be adjusted." If your file carries C2PA metadata from the AI tool that made it, the label stays. Removing a label from a film that really is photoreal AI also defeats the purpose; correct the description instead.

Where is the altered content setting in the YouTube app?

YouTube's help page uses the same steps for the mobile app and a computer: during upload, open the details, find the Attributes section and, under "AI use," tap Yes if the content meets the disclosure requirements. The page doesn't describe a separate Shorts camera flow, so check the details screen before you post. If you can't find it in the app, upload or edit the video in YouTube Studio on a computer.