LinkedIn for tagging images and videos generated by artificial intelligence


While images or videos are generated or faked using generative AI tools they attack social networks few months, LinkedIn announced, in a blog post, implementing specific markup for this type of content. In order to achieve this, the professional social network joined forces with Coalition for the origin and authenticity of content (C2PA): a consortium of technology players that “develops technical standards for certifying the source and origin of media content”, including an invisible marker. It’s called Content Credentials (CR) and it’s already adopted other platforms like TikTokit is the label that will be used on LinkedIn to identify content that has been generated or modified by AI.

A tag that tracks the origin and changes made to the content

Founded by Adobe, the BBC, Intel and Microsoft, LinkedIn’s parent company, the Coalition for Content Provenance and Authenticity (C2PA) has been joined in recent months by several dozen players, including Google, OpenAI and Meta. The organization, which has existed since 2021, designed an invisible marker called Content Credentials (CR), including metadata that allows the identification of the origin of an image or video. And thus ensure that it is authentic or credible. Shown by the body as a “list of ingredients” associated with each piece of content, a Content Credential (CR) tag includes information about the author, creation date, and tools used.

Therefore, any image or video designed or edited using generative AI tools – and which includes C2PA metadata – will have a Cr icon on LinkedIn. By clicking on the pictogram that will be displayed in the upper left corner (see the image above), “you will be able to trace the origin of AI-generated media, including the source and history of the content, and determine whether it was created or modified by AI”, explains Patrick Corrigan, vice president of LinkedIn. This marker will first be placed on images or videos that appear in the news feed, and then spread to ads that are broadcast on the professional social network.

By enabling the traceability of content origin and changes made, C2PA helps maintain the reliability of digital information, protect against unauthorized use and create a transparent and secure environment for creators, publishers and usersLinkedIn rejoices.

Identifying AI-Generated Content: A Big Challenge for Platforms

LinkedIn is not the first platform to try to impose a form of regulation on AI-generated or edited content on its users. In November 2023. YouTube has introduced a number of measures regulate the use of generative artificial intelligence, including the use of special mention. To limit the spread of misinformation, TikTok has also decided impose a clear indication of the use of AI. The Meta group, for its part, is now has already watermarked images created by Imagine with Metaits image-generating artificial intelligence was launched in the United States.





Source link

Leave a Comment