YouTube is putting AI characters where you can see them

After Google expanding its AI validation efforts At I/O, YouTube will now start taking AI content seriously. YouTube has announced that it is moving AI disclosures on shorts and long-form videos to make them easier to watch and will begin identifying and documenting AI-generated content on the platform.

For regular YouTube videos, the logo – which says “AI” next to the famous logo – will appear below the video player, above the description. Currently, this information is hidden on the videos themselves and can be viewed by enhancing video description and look under the “How this was made” section, which requires people to review the details of each video.

For YouTube Shorts, the same AI icon will also appear as a video overlay – apparently YouTube has been there. trying to change about this sign for a while. It also used overlays on clips that flagged whether a video contained “edited or manipulated content.”

I’ve shown that YouTube’s AI system writes it has been inconsistent so farHopefully these changes will establish a framework that the platform adheres to.

“By moving this content to a larger platform, viewers get what they need in less time,” YouTube said in its announcement. “This is now one version of all edited or AI-generated images on YouTube. For content that isn’t real, lifelike, or slightly edited, viewers can find this by specifying the enhanced content.”

YouTube is also expanding its AI writing services by… looking for more AI. The video streaming platform is said to be rolling out “new internal signals” sometime this month that will help it identify and post AI-generated videos. YouTube says it still requires creators to manually disclose when they use AI graphics, but now if a creator doesn’t say whether they used AI or not, the AI ​​logo will be used if YouTube’s system “detects the use of AI graphics.”

If a video has an incorrect profile and is labeled by YouTube, creators can change the YouTube Studio interface. That said, if the creator used YouTube’s AI tools like Veo or Dream Screen, or if the content contains C2PA metadata that indicates it was created by AI, the AI ​​disclosure will be permanent.

YouTube already has automated systems to identify AI-generated content and edits using tags such as C2PA and Google’s SynthID. So I think it’s now committed to doing a better job, especially for AI videos that are trying to mimic real people and environments. YouTube also says that the changes aim to “make it as easy as possible for creators and viewers to have the right information” and that the disclosing labels will not affect monetization algorithms or recommendations.