YouTube Attacks Deepfakes With New AI Policy

YouTube has made changes to its policy on artificial intelligence-generated content. After announcing a strategy to label videos with synthetic content, the tech company has added a new section to its privacy policy. Users will now be able to report deepfakes which attack his personality.

According to YouTube’s privacy policy website, a person can Report AI-generated content that imitates your appearance or voice. This means you will have a removal tool at hand. deepfake created with the help of artificial intelligence, although there is a requirement. The content must show an altered or synthetic version of oneself, realistic.

As with other reports, YouTube will analyze a number of features before deciding whether to remove them. deepfake. Platform evaluates whether the content is modified or synthetic and whether it is shown to the userFake videos will be assessed from several points of view, starting with their realism and the ability to clearly identify the person who reported them.

YouTube will also consider whether the content depicts public or famous figures involved in criminal or violent activity or promoting politicians. A notable point is related to parodies, satires or content of public interest. Let us remember that the latter usually matters when making a decision.

In case the AI ​​generated video violates the rules, The author will have up to 48 hours to edit the video. Claims for synthetic content are similar to those for personal data, so creators can blur, crop, or remove your face. YouTube warns that If the violation continues, additional measures will be taken against the channel..

YouTube Says It’s OK With AI

One of the peculiarities of this rule adjustment is that YouTube implemented it without much fuss. Contrary to its general policy of identifying and removing false content, the platform prefers that affected users directly report the violation. The worst part is to point out deepfake does not always result in video deletion.

YouTubelike many other companies, is betting big on artificial intelligence. A few days ago, it was reported that the company is in talks with three of the most important companies in the music industry. YouTube plans to launch an AI-powered music generator to integrate into its short videos (Shorts).

“Generative AI has the potential to unlock creativity and transform the experience of viewers and creators on our platform,” said Jennifer Flannery, YouTube’s product manager. “We’re in the early stages of our work and will continue to evolve our approach as we learn more.”

The privacy policy changes come as Meta overhauls its synthetic content labeling tool. Starting today, AI-generated images will no longer be labeled as “Generated with AI” and will show more context thanks to a new identifier.