TikTok, the app that was once just a place to store user-generated content, has launched a new setting that allows users to choose how much AI-generated content appears in their “For You” feed. The company is also introducing more advanced labeling technology for AI-generated content.
New AI-generated content (AIGC) controls are being rolled out within the app’s ‘Topic Management’ tool, allowing users to choose what they want to see on TikTok.
“Topic controls let you adjust how often you see content related to more than 10 categories, including dance, sports, food and drink, and more,” TikTok said in a blog post. “Like these controls, the AIGC settings are intended to allow users to customize different content in their feeds, rather than completely removing or replacing content in their feeds.”
The move comes as companies like OpenAI and Meta embrace AI-only feeds. In September, Meta launched Vibes, a new feed for sharing and creating short AI-generated videos. Days after launching Meta, OpenAI released Sora, a social media platform for creating and sharing AI-generated videos.
Since the launch of Sora, realistic AI-generated videos have been posted on TikTok. Additionally, many TikTok users are leveraging AI to create visuals for posts about other topics, such as history and celebrities.
TikTok says new AI-generated content controls will allow users who want to see less of this type of content to do so, while users who enjoy it can choose to see more.
To access the new feature, go to (Settings), select (Content Settings), and click the (Manage Topics) option. You can then move sliders for different topics, such as AI-generated content, to adjust how much or not you want to see that type of content in your For You feed.
TikTok says this change will be rolled out in the coming weeks.
To improve its ability to label AI-generated content, TikTok is currently testing a technology called “invisible watermarking.”
TikTok already requires people to label realistic AI-generated content, using a cross-industry technology called C2PA’s Content Credentials. This embeds metadata in content that lets content and other platforms know that the content was generated by AI. However, TikTok notes that these labels may be removed if the content is re-uploaded or edited on other platforms.
The new “Invisible Watermark” adds another layer of protection with a watermark that only TikTok can read. This means it will be harder for others to remove.
TikTok will begin adding an invisible watermark to AI-generated content created with TikTok tools such as AI Editor Pro. It also adds them to content uploaded using C2PA’s content credentials. The company says these watermarks help make content labeling more reliable. TikTok says it will continue to read C2PA’s content credentials and add them to AI-generated content created on its platform.
In conjunction with these efforts, TikTok also announced the launch of a $2 million AI Literacy Fund targeted at experts like nonprofit Girls Who Code who create content that teaches people about AI literacy and safety.
