YouTube has just announced a major update to its content processing policy created by artificial intelligence (AI), which adds a system to automatically detect and label AI videos with high authenticity.
According to YouTube, this move is aimed at increasing transparency as AI video creation tools are increasingly developing strongly and difficult to distinguish from real images.
The new system will focus on videos that are altered or intentionally created by AI in a way that may mislead viewers. Content belonging to this category will be labeled even if the creator does not self-declare the use of AI.
Previously, YouTube required creators to proactively disclose if the video uses AI to simulate images, voices or realistic elements. However, the company said they are now able to automatically detect some types of AI content and proactively label when necessary.
For long videos, notifications about the use of AI will appear right below the video player and above the description. Meanwhile, with YouTube Shorts, the AI label will appear directly as a covering on the video.
YouTube said the stricter labeling system is mainly applied to authentic AI videos such as real-life simulation scenes or real-life events. Animated videos, unrealistic content, or simple editing will still use a lighter notification form in the expanded description.
Some AI labels will not be able to be manually removed. Videos created using YouTube's AI tools such as Veo or Dream Screen, or videos containing C2PA metadata confirming content created by AI, will always retain the AI recognition label.
If you think your video is labeled incorrectly, the creator can send a review request through YouTube Studio.
In addition to changes in AI policy, YouTube also introduced a new personalized content source system. This feature allows users to create a list of videos according to their own preferences by entering descriptions of the content they want to watch.
For example, users can ask the video suggestion system for mood, favorite topic or specific type of content. YouTube will use view and search history to build a feed that is more suitable for each individual.
According to the company, this feature has been tested since the end of 2025 and is now being deployed for users who log in in the US on both mobile applications and computer versions. To use it, users need to turn on search history and video view history.
