The test
On 18 March 2024, YouTube announced a Creator Studio tool requiring disclosure when a video's realistic content is altered or synthetically generated. The original announcement states YouTube was 'introducing a new tool in Creator Studio requiring creators to disclose to viewers when realistic content, content a viewer could easily mistake for a real person, place, scene, or event, is made with altered or synthetic media, including generative AI'. A later post, Improving AI labels for viewers and creators, dated 27 May 2026, revisits the same rule to change how the resulting label is shown.
What the evidence says
The 2024 post is YouTube's own policy launch, listing the disclosures required: replacing a person's face, synthetically generating their voice, altering real footage to misrepresent events or places, or depicting a realistic fictional event at a real location, with exemptions for clearly unrealistic content, productivity uses like automatic captions, and minor changes such as color filters. At launch, most labeled videos displayed the disclosure 'in the expanded description', with more prominent on-player labels reserved for sensitive topics such as health, news, elections or finance. The 2026 post changes that placement for everyone: for long-form video, labels now appear 'directly below the video player, above the description', and for Shorts, 'as an overlay on the video itself', while less significant alterations still disclose only in the expanded description. Read together, the posts show a documented change in label visibility rather than in which content requires disclosure.
The sample and the variance
Both are YouTube's own blog posts describing its own product; neither is an independent measurement of how many videos are labeled or how often creators disclose voluntarily versus have a label added by YouTube itself. The 2024 post is dated at publication; the labeling rule it describes functions as a living policy the 2026 post itself revises, so a description of label placement should specify which date it reflects.
What to try next
As an editorial reading, a creator altering a real person's likeness, voice or a real event's footage should apply YouTube's disclosure tool regardless of era, since the 2024 disclosure categories do not appear to have narrowed in the 2026 update. A publisher citing this policy should match the placement rule to the date in question, since 'in the description' was accurate at launch but 'below the player or as an overlay' is the current standard.
- Does the content meet the 2024 disclosure categories: face replacement, synthetic voice, or misrepresented real events or places?
- Is the video long-form or a Short, since the 2026 update places the label differently for each format?
- Does the topic fall into a sensitive category, such as health, news, elections or finance, that both posts treat with stricter label prominence?
The record establishes a documented sequence: a March 2024 disclosure requirement with description-level labels for most content, followed by a May 2026 change that made those labels more visible without, on the evidence available, changing which content must be disclosed in the first place.
Sources & limits
- Disclosing AI-generated content for added transparency ↗
Announces the original 2024 disclosure requirement, its categories, exemptions, and description-level label placement at launch.
Source · Source date: 2024-03-18 · Archive retrieval: 2026-09-16 - Improving AI labels for viewers and creators ↗
States the May 2026 change to more visible label placement (below the player for long-form, an overlay for Shorts).
Source · Source date: 2026-05-27 · Archive retrieval: 2026-09-16