Deepfake Defense: YouTube Introduces Policy to Remove AI-Made Content

YouTube Introduces Policy to Remove AI-Made Content

Meta isn’t alone in grappling with the rise of AI-generated content. In June, YouTube updated its policies to remove videos using AI to create fake faces or voices. This change is part of YouTube’s existing privacy request process and builds on their commitment to responsible AI use, which they announced last November.

YouTube now handles takedown requests for AI-generated content differently from misleading content like deepfakes. In the past, viewers might have reported these videos as deceptive. Now, YouTube requires individuals directly affected by AI-generated fakes to request removal under privacy violation rules.

Want a Free Website

According to YouTube’s updated help guide, in most cases, the person depicted in the video must submit the takedown request themselves. Exceptions apply to minors, individuals unable to access a computer, deceased persons, or other special circumstances.

You may also like to read: YouTube’s Latest Feature Lets Music Partners Take Down AI Deepfakes. 

After receiving a takedown request, YouTube does not automatically remove the content. They evaluate several factors before deciding, such as whether the video discloses it’s AI-generated, its context (e.g., parody, satire, newsworthy), and the portrayal of public figures (e.g., criminal activity, violence, endorsements).

YouTube grants video creators 48 hours to respond to a removal request. The matter is resolved if the video is taken down within that timeframe; otherwise, a formal review process begins.

YouTube clarifies that “removal” means removing the video and scrubbing the person’s name and personal details from the title, description, and tags. It isn’t enough to make the video private because it might subsequently be made public again. Although blurring faces can address concerns, a complete removal must comply with a takedown request.

Want a Free Website