YouTube has just activated a new and highly visible AI labeling system. As of May 27, 2026, the platform now places a prominent AI generated content label directly under the video player for long form videos and as an on screen overlay for Shorts. This is a major change. The label is no longer buried in the description. It is now unavoidable and placed in the same visual position YouTube uses for safety notices. The Help Center confirms that the purpose is to alert viewers when content may not depict real people, real events, or real places.
Primary sources:
YouTube Help Center: About AI generated content labels
https://support.google.com/youtube/answer/15447836
YouTube Official Blog (May 27, 2026)
https://blog.youtube/news-and-events/ai-generated-content-disclosures-update/ (blog.youtube in Bing)
General AI Label Policy
https://support.google.com/youtube/answer/13822102 (support.google.com in Bing)
This raises immediate legal questions.
• Is the rule being applied uniformly? If the standard is that content may not depict real people or real events, then nearly all modern productions qualify.
• Does CG equal AI? Films like Remarkably Bright Creatures openly disclose CG, but many CG pipelines now use machine learning tools. If ML is used anywhere in the process, does that trigger the label.
• How can this be enforced consistently? The detection system is opaque. Two creators using the same tools may be treated differently depending on metadata or classifier behavior.
• Does it matter? We may be witnessing the emergence of a new medium. The creator profiled in the TMZ article from May 24, 2026, titled “First Time in San Juan,” is earning significant revenue from AI generated music and visuals. His audience embraces the work. The market may care more about authenticity of intent than whether the pixels were created by a model or a camera.
TMZ reference:
https://www.tmz.com/2026/05/24/first-time-in-san-juan-viral-song-explained/
Copyright questions also remain open.
• Thaler v. Perlmutter (D.D.C. 2023) held that AI generated output cannot qualify as human authorship. This reinforces the legal distinction between synthetic and human created works.
• Naruto v. Slater (9th Cir. 2018) confirms that rights attach to human creative acts. This becomes relevant when platforms notify viewers that content may not depict real human activity.
• At the same time, the TMZ example shows that audience acceptance and monetization can occur regardless of authorship status. In some commercial contexts, ownership may matter less than distribution, branding, and engagement.
Bottom line: The new, prominent AI label is a significant shift in how YouTube frames synthetic media. Whether it functions as a true warning, whether it can be applied uniformly, and whether copyright ownership remains central in this new medium are all open questions that directly affect our practice.