YouTube is increasing the visibility of its labels for AI-generated content and is also implementing automatic detection of highly realistic AI-generated images and videos. These updates come in response to user feedback and were announced on the YouTube blog.
The platform is also working on how it labels long-form and short-form videos, adding internal signals that can automatically apply labels without requiring input from creators.
Where do YouTube’s new AI labels appear and how does automatic detection work?
The location of the new label varies depending on the video format. For long-form videos, the AI disclosure label is now located directly below the video player and above the description. In contrast, for Shorts, the label appears as an overlay directly over the video.
These labels apply to photorealistic content as well as content that has been meaningfully altered or generated by AI. Although YouTube describes the new positions as more noticeable, the labels in the platform’s own promotional videos are still relatively small and may be easy to overlook during regular viewing.
The main change is the introduction of new internal signals that help YouTube automatically recognize AI-generated content. Currently creators are required to manually add reveals when using realistic AI, but not everyone does this consistently.
The new detection technology allows YouTube to automatically label content that contains significant photorealistic AI use. Creators still have some control over these labels.
If they believe their content has been incorrectly identified as AI-generated, they can update their disclosure status in YouTube Studio.
When AI labels can’t be removed and why this update matters
YouTube specifies that in some situations, the disclosure will remain permanent regardless of what the creator wishes. This includes content created with YouTube’s own AI tools such as Veo or Dream Screen, and content that includes C2PA metadata indicating that it was generated entirely by AI.
In these cases, the label cannot be removed through YouTube Studio. The changes come amid growing concerns over the amount of AI-generated content on YouTube. A recent study found that more than 20 percent of videos shown to new users were AI-generated low-quality content.
Separate research shows that as technology improves, it is becoming increasingly difficult for viewers to distinguish AI-generated videos from real footage.
The label update is part of several recent YouTube efforts related to AI, including expanding similarity detection to all creators 18 and older and launching Gemini Omni for shorts remixing.
YouTube has not specified when the re-labeling and automatic detection will be fully available to all users and regions.
