New Update: TikTok Users Can Now Adjust How Much AI Content They See
TikTok, an app once known primarily as a destination for user-generated content, is now launching a new setting that allows users to choose how much AI-generated content (AIGC) they want to see in their "For You" feed. This new AIGC control is being launched within the "Manage Topics" tool, which was previously used to adjust the frequency of topics such as Sports or Food. TikTok explains that this setting aims to help users tailor the variety of content in their feed, rather than removing or replacing the content entirely.
This move comes as other tech giants like OpenAI and Meta are beginning to embrace the concept of AI-specific feeds. In September, Meta released "Vibes," a new feed for sharing AI-generated short videos, while OpenAI launched "Sora," a social media platform for AI-generated videos. Since Sora's arrival, realistic AI videos have increasingly appeared on TikTok, in addition to users leveraging AI to create visuals related to history or celebrities. With this new feature, users who find such content bothersome can reduce its presence, while those who enjoy it can choose to see more.
To access this new capability, users can navigate to the menu: Settings > Content Preferences > Manage Topics. There, users can move the slider on various topics, including AI-generated content, to adjust display preferences in the "For You" feed. TikTok confirms that this change will start rolling out to all users in the coming weeks.
Beyond user controls, TikTok is also introducing a more advanced labeling technology called "invisible watermarking." Previously, TikTok utilized an industry technology called Content Credentials from C2PA, which embeds metadata to detect AI content. However, these labels often disappear if the content is re-uploaded or edited on other platforms. With "invisible watermarks," TikTok adds a security layer in the form of a watermark that can only be read by their own system, making it much harder for third parties to remove or manipulate.
TikTok will begin applying these invisible watermarks to content created using internal tools like AI Editor Pro, as well as content uploaded with Content Credentials from C2PA. The company states that this method will make content labeling more reliable. In line with these technical efforts, TikTok also announced the launch of a $2 million AI literacy fund aimed at experts and non-profit organizations, such as Girls Who Code, to create educational content teaching the public about AI literacy and safety.