YouTube has updated its privacy guidelines, allowing users to report artificial intelligence (AI) generated content that looks or sounds like them (deepfakes). This doesn’t mean that just any instance of someone identifying themselves in a video would be considered for removal. When evaluating the content, YouTube will consider factors such as:

• Whether the creators have disclosed that the content is altered or synthetic
• Whether the person whose likeness has been replicated is uniquely identifiable. This means that there should be enough information about you in the video for others to recognize you.
• Whether the content is realistic (Note: the company has not explained what qualifies as realistic content).
• Whether the content contains parody, satire, or other public interest value
• Whether the content features a public figure or well-known individual engaging in sensitive behavior such as criminal activity, violence, or endorsing a product or political candidate

The creator would have 48 hours to act on the complaint during which the creator can trim or blur out the section featuring the deepfake. Creators can also choose to take down the video completely which will close the complaint. YouTube will act on the complaint only if the creator doesn’t take the necessary action.

Some context:

This update comes after the company announced that it would allow people to submit removal requests for AI-generated or other synthetic or altered content that simulates an identifiable individual in November last year. As a part of its updated policy, the company had said that creators would be required to label pieces of altered or synthetic content that look realistic, and creators not labeling such content would be subject to penalties such as content removal and suspension from the YouTube Partner Program. 

The company has notably also updated its policy about the use of AI-generated content featuring victims of crime in January this year, saying that such content will be removed from the platform. 

Also read: