TikTok to label AI-generated content

TikTok will start labeling artificial intelligence-generated content uploaded from other platforms as part of a push to tackle misinformation, the company said Thursday. 

TikTok is partnering with the Coalition for Content Provenance and Authenticity (C2PA), a project that supplies resources necessary to identify AI-generated content. 

In the announcement, TikTok said it will use C2PA’s “Content Credentials” technology, which attaches metadata to content that indicates to users it is AI-generated. The feature is available for both images and videos starting Thursday but will be coming “soon” for audio-only content, according to TikTok. 

The decision is part of a broader effort by technology platforms to increase transparency of AI usage on their platforms. Meta, the parent company of Facebook and Instagram, said in February it will start labeling more AI-generated content and remove some that violate its policies. 

Increased labeling of AI-generated content was included in President Biden’s executive order signed last October, which focused on managing the risks of emerging technology while ensuring safety, privacy, and innovation. 

TikTok already labels content made with tools inside the app. 

Apart from watermarking content, TikTok said it will be releasing 12 videos that “highlight” media literacy skills. TikTok will work with MediaWise, Poynter Institute’s program focused on misinformation. 

“Our Teen Fact-Checking Network has built an audience with innovative media literacy videos on TikTok since 2019. Five years later, we’re thrilled to empower even more people to separate fact from fiction online,” MediaWise’s Director Alex Mahadevan said in a statement.