TikTok lets users control AI-generated content in feeds while expanding AI labeling and literacy efforts

TikTok is introducing a feature that lets users adjust how much AI-generated content appears in their feeds. The company reports it already hosts over 1 billion AI videos and plans a broader rollout after testing in the coming weeks. The update comes amid a surge in AI-created content from tools like OpenAI’s Sora and Google’s Veo 3. About 1.3 billion videos on TikTok are labeled as AI-generated, with overall uploads exceeding 100 million per day, making AI content a relatively small portion of the platform. Users can alter AI content exposure via the “manage topic” setting by selecting “AI-generated content,” alongside existing controls for topics like current affairs, fashion, beauty, and dance. TikTok is also adding an “AI-made” watermark to content produced with its AI tools or identified by the industry-wide C2PA initiative, aiming to deter attempts to evade labeling. The platform is launching a $2 million AI literacy fund to support organizations like Girls Who Code in producing educational content about responsible AI use. Separately, TikTok faces scrutiny over trust and safety moderator layoffs in the UK, with plans to cut 439 roles in London. Company executives emphasize that human moderation remains essential and that AI assists in reducing exposure to harmful content, recommending a balanced approach between automation and human oversight. Overall, TikTok highlights a notable decrease in shocking content seen by moderators, attributed to automated systems.