TikTok has announced that it will begin labeling images and videos uploaded to its platform that have been generated by artificial intelligence (AI), employing a digital watermark known as Content Credentials. This move aims to enhance transparency and combat misinformation as AI-generated content becomes more prevalent.
The social media giant already labels AI-generated content created within its app. However, the latest policy extends this practice to include content generated by external AI tools. “We have policies that prohibit realistic AI that is not labeled,” said Adam Presser, head of operations and trust and safety at TikTok, in an interview. “If realistic AI-generated content appears on the platform without a label, we will remove it as violating our community guidelines.”
Researchers have raised concerns that AI-generated content could be used to interfere with elections and spread misinformation. TikTok’s shift aligns with a broader industry effort to provide safeguards for AI usage. In February, Meta announced collaborations on technical standards to identify AI-generated images, videos, and audio, with plans to label such content on Facebook and Instagram.
The Content Credentials technology was developed by the Coalition for Content Provenance and Authenticity, co-founded by Adobe, Microsoft, and others. The technology attaches metadata to content, allowing platforms to recognize and label AI-generated material instantly. “TikTok is the first social media platform to support Content Credentials,” said Dana Rao, Adobe’s executive vice president, general counsel, and chief trust officer. “Their platform and vast community of creators are essential in increasing transparency online.”
For the system to function effectively, both the creators of AI tools and content distribution platforms must adopt the industry standard. When a user generates an image with OpenAI’s DALL·E tool, for example, OpenAI attaches a watermark and metadata to the file. If that image is uploaded to TikTok, it will automatically be labeled as AI-generated.
TikTok, owned by ByteDance, has also joined the Adobe-led Content Authenticity Initiative to promote the adoption of Content Credentials across the industry. The company began deploying the technology on Thursday for images and videos, with plans to extend it to audio content soon.
In related developments, TikTok and ByteDance have filed a lawsuit to block a U.S. law that would require ByteDance to divest TikTok or face a ban, arguing that it violates the First Amendment. TikTok continues to expand its user base, boasting over 170 million users in the United States alone.
Reference(s):
TikTok to label AI-generated content from OpenAI and elsewhere
cgtn.com
