source: techcrunch ai: youtube will now automatically label ai videos
level: business
youtube is shifting from relying on creators to self-report ai use to automatically labeling videos when its systems detect significant photorealistic ai. the company announced that internal signals will identify such content and apply labels, starting in may. creators can still disclose ai use manually, but if they skip it, youtube will label the video for them. however, labels cannot be removed if the content was made with youtube's own ai tools like veo or dream screen, or if it carries c2pa metadata showing full ai generation.
the labels are also becoming more visible. for long-form videos, they will appear directly below the player above the description; for shorts, they will overlay on the video. this change aims to make it easier for viewers to spot ai-altered or generated content. videos with only minor ai edits or clearly animated scenes will keep labels in the expanded description only. youtube says these labels won't affect video recommendations or monetization.
this update follows google's release of gemini omni, a multimodal ai model that produces high-quality video. youtube has been using ai labels for over two years, but the new automatic detection adds enforcement. it also comes after youtube expanded its ai deepfake detection, letting any adult scan for face matches. the company continues to invest in ai features like interactive search and video summaries.
why it matters: automatic labeling helps viewers identify ai-generated content more reliably, reducing misinformation risks without relying on creator honesty.
source: techcrunch ai: youtube will now automatically label ai videos