The Chinese social media platform has created a tool to automatically identify AI-generated content. It is part of the platform’s efforts to “counter and identify AI-generated content” in the app.
About a year ago, TikTok started requiring all users to label their content if it is AI-generated. This is to avoid confusing or misleading users if they don’t know that the content they are watching was generated by AI, thus making the “context clear”. To make it easier for users, it also launched a dedicated tool within the platform for this purpose.
Now, the platform has launched a new tool to automatically identify AI-generated content when it is uploaded from certain other platforms. This has been done in collaboration with the Coalition for Content Provenance and Authenticity (C2PA) project, which aims to counter misleading information online.
It is within this project that a new technique for identifying AI content was found. TikTok will be the first video sharing platform to use this technology where metadata is attached to content, allowing it to recognize if the content is AI-generated.
“Crucial step”
Furthermore, the platform joins the Content Authenticity Initiative (CAI), which is a “group of creators, technologists, journalists and activists leading the global effort to address digital misinformation and content authenticity” and is led by Adobe. Among others, the BBC, Microsoft, Intel and photography company Nikon are part of the collaboration.
– At a time when all digital content can be manipulated and tweaked, providing ways for the public to distinguish between what is disinformation and what is not is crucial for authenticity. Today’s announcement is a critical step towards achieving that outcome, said Dana Rao, General Counsel and Chief Trust Officer at Adobe, in a press release.
Related articles:
When it comes to legislation on AI-generated porno...
