YouTube to Start Requiring Content Creators to Label AI-Generated Videos

YouTube will be starting to require content creators to disclose "altered or synthetic" realistic-looking AI-generated media.

In a blog post on Monday, the video-streaming giant announced that it will be rolling out a new tool on YouTube studios to allow content creators to notify their viewers that a video is AI-generated through the expanded description.

The new tool will work on an honesty system, asking content creators whether the video they will be uploading is AI-generated or not.

However, in the case a channel fails to disclose that its content is AI-generated, YouTube will still be able to flag the video once a viewer reports the content to be unlabeled.

Repeat offenders will have their contents at risk of removal or even be demonetized.

YouTube AI Labels Will Only Apply to Realistic Looking Videos

It is worth noting that the disclosure is only required for videos and media that either use the likeliness of an actual person, altering real events and places, or realistic depiction of "fictional major events."

AI-generated content that is unrealistic or only has "inconsequential" minimal changes will not be required for a disclosure. This includes AI-generated animations, color adjustments, special effects, and beauty filters.

Content creators will not also be required to disclose if they used generative AI for scripts, content ideas, and automatic captions.

YouTube's plan to launch its own AI-generated content label was first announced in November last year as part of its efforts to "help increase transparency around digital content."

Also Read : Meta Vows to Detect AI-Generated Images More Ahead of 2024 Elections