Photo Credit: Getty Images
TikTok is rolling out a new control that lets users decide how much AI-generated content appears on their For You feeds, marking the platform's strongest attempt yet to manage the surge of synthetic media dominating short-form video.
The feature, which will be tested in the coming weeks before launching globally, arrives as powerful video-generation tools, including OpenAI's Sora and Google's Veo 3, flood the internet with machine-made clips. TikTok says its platform already hosts more than 1.3 billion AI-labeled videos, though AI content still represents a small fraction of the 100 million uploads posted each day.
Jade Nester, TikTok's European director of public policy for safety and privacy, said the shift is meant to give users more agency. "Many people enjoy content made with AI tools, from digital art to science explainers," she said. "We want to give people the power to see more or less of that, based on their own preferences."
The new setting will sit under TikTok's "Manage Topics" menu, where users already adjust categories such as fashion, current affairs, beauty, and dance. The slider for "AI-generated content" will allow viewers to increase or reduce the presence of synthetic media without eliminating it entirely.
The move comes as AI videos grow more realistic, and harder to detect. TikTok requires creators to label lifelike AI visuals and bans harmful deepfakes involving public figures or crisis events. To strengthen enforcement, TikTok is introducing invisible watermarking, a new layer of digital tracing that persists even if clips are edited or reuploaded. It will apply to videos created with TikTok's in-app AI tools as well as content flagged through C2PA metadata.
TikTok is also launching a $2 million AI literacy fund, partnering with groups such as Girls Who Code to teach responsible, safe use of AI across digital communities.
The company defended its strategy as it faces criticism over job cuts that will affect hundreds of UK-based moderators. Brie Pegum, TikTok's global head of program management for trust and safety, said human judgment remains critical but stressed that AI systems have reduced moderators' exposure to graphic content by 76% in the past year.
