YouTube now automatically applies AI labels to videos detected with "significant photorealistic AI," shifting responsibility from creators, according to TechCrunch on May 27, 2026. This move enhances transparency for viewers across long-form content and Shorts by making disclosures more prominent and enforced by the platform.
For over two years, YouTube has required creators to self-disclose AI content, particularly when it could be mistaken for real people, places, or events. However, the rise of advanced AI video models has led the platform to take a more proactive stance.
Why Is YouTube Automating AI Labeling?
YouTube is automating AI labeling to combat the increasing prevalence of highly realistic AI-generated content and to ensure viewers are properly informed. This shift aligns with the growing capabilities of generative AI tools that can produce convincing video content.The company stated it would now use new internal signals to identify AI-generated content. This follows the recent unveiling of Google's Gemini Omni multimodal AI models, which can create high-quality videos demonstrating an understanding of physics and cultural nuances. While creators still need to disclose AI use, YouTube will apply labels automatically if they do not.
Labels will be permanently attached to videos containing C2PA metadata, indicating full AI generation, or if the content was created using YouTube's own AI tools like Veo or Dream Screen. Creators cannot remove these specific labels.
| Labeling Scenario | Creator Disclosure | YouTube Automation | Label Placement |
|---|---|---|---|
| Photorealistic AI | Required (if missed, then automated) | Yes | Below video player/On Shorts |
| Animated/Unrealistic AI | Optional (e.g., prancing unicorn) | No | Expanded description only |
| Content from YouTube AI Tools | Not applicable | Permanent | Below video player/On Shorts, unremovable |
| Content with C2PA Metadata | Not applicable | Permanent | Below video player/On Shorts, unremovable |
How Will This Impact Creators and Viewers?
This new policy provides viewers with clearer, more consistent information about AI-generated content while streamlining the labeling process. Viewers will now find these disclosures in more prominent locations, making it easier to distinguish between human-made and AI-made videos.Previously, AI labels were often hidden in the expanded description, appearing directly on the video only for sensitive topics like health or news. Now, for long-form videos, labels for photorealistic or significantly altered AI content will appear directly below the video player. On YouTube Shorts, these labels will overlay the content itself, according to CNET.
YouTube confirms these AI labels will not influence a video's recommendation status or its ability to generate revenue. This ensures creators are not penalized for using AI, only for failing to disclose it. The platform also expanded its AI deepfake detection, allowing any adult to scan for face matches, building on earlier tests with public figures and celebrities.
