Does a realistic AI product video need a YouTube disclosure?

Describe how the finished scene was created

Imagine an illustrative campaign showing a real travel mug on a moving train table. The merchant supplied a product photo, but the location and motion were generated. A viewer could take the clip for footage shot on an actual journey. The existence of the mug does not establish that the depicted event occurred, so record the source and generated contribution separately.

List the actual ingredients of the export: real footage, generated imagery, motion generation, editorial text and any added audio. An externally edited video may combine several sources. Make this record about the published result rather than simply naming the tool used or saying the entire campaign involved AI.

Compare the scene with the official disclosure criteria

YouTube’s current GenAI disclosure guidance requires disclosure for realistic content meaningfully generated or altered with AI, including a realistic scene that did not occur. It distinguishes those uses from non-realistic content, minor aesthetic changes and some production assistance. Its examples guide the decision but are not an exhaustive list.

For the mug, apply the realistic-scene criterion to the generated train setting and movement. A separate case involving only AI-assisted script planning would require a different assessment. Do not assume every small use of an AI tool creates the same disclosure obligation, or that a real product photo makes all later generated footage a recorded event.

Preserve a scene-level decision record

Record uncertainty rather than resolving it through an attractive thumbnail. Ask the publisher to review the final combination if the scene or narration changes. This is a merchant editorial record, not an automatic classification service or a legal determination for all jurisdictions.

Fictional mug release M6 has three candidates. A shows a realistic train journey created from a still photo; that journey was never filmed. B only adjusts the color of genuine footage without materially changing the depicted event. C used AI to plan a script but publishes unaltered footage. A falls within the realistic generated-scene criterion; B and C need their actual edit scope assessed against the minor-edit and production-assistance examples. Do not give all three the same answer solely because AI appeared somewhere in the workflow.

Record the exact export and the basis for the choice. If an editor later adds a realistic generated passenger, changes what the mug appears to do, or adds synthetic speech, reopen the assessment for that version. An automatic label or a manual disclosure does not make a leak-resistance claim true. These candidates are teaching examples; no YouTube control, upload or generated file was tested.

Use the destination’s actual upload control

As checked on October 7, 2026, YouTube’s upload guidance describes an AI use setting in the upload Attributes section. Select the appropriate response using the current interface and applicable criteria. Older instructions may call the setting altered content; use the actual account interface and current official help when labels differ.

Review the published presentation and keep the upload decision with the URL and export version. In this publication record, the authorized publisher must check the actual setting and labels received by YouTube. A source-tool name or a local note is not proof that the platform received a disclosure. A descriptive note in your own records should support the publisher’s decision, not be treated as proof that another platform received the disclosure.

Continue the product-fact review after disclosure

Inspect the file from the route and model actually used, including any generated or added audio. Review every scene and the complete edit for the product’s shape, labeling and visible behavior. A platform AI label does not establish that the mug resists leaks or performs on a train as shown. Obtain real evidence for any such product claim.

Use the UGC-versus-testimonial guide if an added voice or review card implies an actual user experience. Use the mixed-footage guide for source relationships. Keep accurate creation context and accurate product claims together; neither replaces the other, and this article does not promise comprehensive advertising or legal compliance.