YouTube confirms it will remove AI fakes within 48 hours if complaints are upheld • The Register

AI Video & Visuals

YouTube has instituted privacy guidelines that allow users to request the removal of AI-generated videos that imitate them.

YouTube has committed to doing so in November 2023, assuring users that the rules will come into effect “in the coming months.” A recent update to YouTube's privacy guidelines suggests that time has come.

“If someone has used AI to modify or create synthetic content that looks or sounds like you, you can ask for it to be removed,” the guidelines currently state. “To be eligible for removal, content must realistically modify or synthesize your likeness. We consider a variety of factors when assessing complaints.”

Here are some things YouTube takes into consideration when reviewing removal requests:

  • Whether the content is altered or synthesized
  • Whether the content is presented to the audience in an altered or synthetic form.
  • Can an individual be uniquely identified?
  • Is the content realistic?
  • Whether the content contains parody, satire, or other public interest value.

Complaints are moderated by humans and, if approved, the uploader has 48 hours to either remove the video entirely or edit out the parts that violate our privacy guidelines by cropping or blurring them. If the deadline expires without the uploader taking action, YouTube will remove the violating content.

On the surface, the rules appear to be a safeguard against inappropriate and malicious content, such as deepfakes, which are easy to create these days. The focus on political endorsements by public figures is also particularly timely, as 2024 is a key election year for many countries, including France, the UK and the US.

YouTube did not respond to a request for comment before publication.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *