TikTok Introduces New Controls for Managing AI-Generated Content in User Feeds

TikTok is expanding its content management tools by allowing users to control how much AI-generated material appears in their “For You” feed. Alongside this feature, the platform is deploying more advanced labeling technologies to better identify AI-created posts.

Customising AI Content Visibility

The new AI-generated content (AIGC) preference is being added to TikTok’s “Manage Topics” section, which already lets users fine-tune the categories they encounter on the app.

According to TikTok,
“Manage Topics already enables people to adjust how often they see content related to over 10 categories like Dance, Sports, and Food & Drinks. Like those controls, the AIGC setting is intended to help people tailor the diverse range of content in their feed, rather than removing or replacing content in feeds entirely.”

This update arrives at a time when major tech companies are leaning heavily into AI-only content feeds. In September, Meta debuted Vibes, an AI-driven feed for creating and sharing short-form AI videos. Shortly after, OpenAI unveiled Sora, its own social media platform centered on AI-generated video creation.

Since Sora’s introduction, the volume of hyper-realistic AI video content shared on TikTok has surged. Many creators have also begun using AI tools to craft visuals for topics such as celebrity commentary, historical summaries and storytelling.

How Users Can Adjust Their AI Content Exposure

TikTok’s new control mechanism allows users to either reduce or increase the amount of AI-generated content they encounter. This provides flexibility for those who prefer traditional user-made videos and those who enjoy exploring AI-created media.

To access the setting:

1. Open Settings

2. Navigate to Content Preferences

3. Select Manage Topics

4. Adjust the slider for AI-generated content

TikTok notes that the feature will be made available to all users in the coming weeks.

Advancing AI Content Labeling with Invisible Watermarks

To strengthen its detection and labeling capabilities, TikTok is testing a new system known as “invisible watermarking.”

Currently, the platform requires creators to label realistic AI-generated content and uses Content Credentials from C2PA, a cross-industry standard for embedding metadata to signal AI involvement. However, TikTok acknowledges that these metadata tags can be stripped when content is edited or reuploaded elsewhere.

Invisible watermarking adds another safeguard. The watermark is embedded in a way that only TikTok’s systems can detect, making it far more difficult to remove.

The company will begin applying invisible watermarks to:

• AI content created using TikTok’s own tools, such as AI Editor Pro

• AI-generated content uploaded with C2PA’s Content Credentials

TikTok says this enhancement will help it label AI content “more reliably”, and it will continue supporting C2PA-based identification across its platform.

TikTok Launches $2 Million Fund for AI Literacy

In a related initiative, TikTok is introducing a $2 million AI literacy fund aimed at supporting organisations and experts working to improve public understanding of artificial intelligence. Beneficiaries will include groups such as Girls Who Code, which will produce educational content focused on AI safety and literacy.

Author: Din Kumar