TikTok has unveiled a new feature that automatically labels AI-generated content originating from other platforms, the company revealed on Thursday. With this update, content creators posting material on TikTok generated via services like OpenAI’s DALL·E 3 will see an “AI-generated” label attached, informing viewers that AI technology was used in its creation.
This initiative is made possible through the adoption of Content Credentials, a technology developed by the Coalition for Content Provenance and Authenticity (C2PA), co-founded by Microsoft and Adobe. Content Credentials embed specific metadata into content, enabling TikTok to promptly recognize and label AI-generated content.
As a result, TikTok will now automatically assign the “AI-generated” label to content uploaded with Content Credentials attached. This update commences rollout on Thursday and will extend to all users globally in the following weeks.
While TikTok already identifies content created with its own AI effects, it will now extend this labeling to content produced on external platforms implementing Content Credentials, such as OpenAI’s DALL·E 3 and Microsoft’s Bing Image Creator. Notably, Microsoft, Adobe, and OpenAI are among the early supporters of Content Credentials, with Google also pledging its support.
Although TikTok already mandates creators to disclose AI-enhanced content, the platform views this update as an additional measure to ensure transparent labeling of AI-generated content while alleviating pressure on creators.
In the near future, TikTok plans to apply Content Credentials to AI-generated content produced on its platform using TikTok AI effects. This metadata will include information on the origin and editing process of AI-generated content, remaining attached even when downloaded. Other platforms adopting Content Credentials will similarly label such content automatically.
While TikTok prioritizes labeling AI content within its own ecosystem, it also aims to ensure accurate labeling when this content is shared on other platforms.
“AI-generated content provides an incredible avenue for creativity, but transparency is paramount for viewers,” stated Adam Presser, Head of Operations and Trust & Safety at TikTok. “Through collaborative labeling efforts across platforms, we empower creators to responsibly explore AI-generated content, while deterring harmful or misleading content that violates TikTok’s policies.”
TikTok emphasizes its distinction as the first video-sharing platform to implement Content Credentials. Additionally, Meta announced plans in February to leverage C2PA’s solution to enhance content provenance.
In its Thursday announcement, TikTok reaffirmed its commitment to combatting deceptive AI use in elections and emphasized its strict policies against harmful or misleading AI-generated content, irrespective of labeling.
