TikTok launched new controls today allowing users to adjust how much AI-generated content appears in their “For You” feeds, marking the platform’s latest move to balance user preferences with the billion-plus AI videos now hosted on the service.
The AI-generated content (AIGC) control integrates into TikTok’s existing Manage Topics tool, which already enables users to customize their feed across over 10 categories including Dance, Sports, and Food & Drinks. Users access the feature through Settings, then Content Preferences, and finally the Manage Topics option. A slider allows them to dial AI content “down” to see less or increase it to see more, though no option exists to block it entirely. TikTok states the control helps users “tailor the diverse range of content in their feed, rather than removing or replacing content in feeds entirely.”
The feature joins TikTok’s broader personalization suite alongside customizable keyword filters and the “not interested” button. The company revealed it currently hosts over one billion AI videos on its platform, underscoring the scale of AI-generated content’s presence across the service.
Image: TikTok
Invisible Watermarks and Content Authentication
TikTok is simultaneously testing “invisible watermarking” technology designed to improve AI content detection and labeling. The watermarks are readable only by TikTok itself, making them significantly harder to remove compared to standard metadata that can be stripped when content is reuploaded or edited on other platforms.
The company will add these invisible watermarks to AI-generated content created with TikTok tools like AI Editor Pro and to content uploaded with C2PA’s Content Credentials. This dual-watermarking approach complements TikTok’s existing use of C2PA’s Content Credentials technology, which embeds metadata identifying AI-generated material. The platform has become the first video sharing platform to implement C2PA’s Content Credentials technology, automatically labeling AI-generated content when uploaded from certain other platforms.
Over the coming months, TikTok will attach Content Credentials to content created on its own platform, with metadata remaining on downloaded content. Anyone can use C2PA’s Verify tool to identify AIGC made on TikTok and learn when, where, and how the content was made or edited. The platform already requires creators to label realistic AI-generated content and uses the cross-industry Content Credentials standard to identify AI-generated material.
TikTok also announced a $2 million AI literacy fund targeting experts and organizations like Girls Who Code to create educational content about responsible AI use and safety. The company launched new media literacy resources developed with guidance from MediaWise and WITNESS to help users navigate AIGC and misinformation online. Additionally, TikTok joined the Adobe-led Content Authenticity Initiative to drive industry-wide adoption of Content Credentials technology.
The new feed controls are rolling out in the coming weeks, while the invisible watermarking technology remains in testing phases. These moves position TikTok among social platforms increasingly addressing AI content as generative tools become more accessible and AI-created material proliferates across networks. The company’s approach combines user agency through content controls with technical infrastructure for content authentication, addressing both consumer preferences and broader industry concerns about transparency in AI-generated media.
