TikTok Now Lets You Control How Much AI-Generated Content Appears in Your Feed

Taking Control of Your TikTok Experience TikTok is rolling out a game-changing feature that puts you in the driver's seat when it comes to AI-generated content. With the new AI-generated content control, users who want to see less of this sort of content can now dial things down, while those who enjoy it can choose to see more of it. This represents a significant shift in how social media platforms approach algorithmic content delivery, acknowledging that not everyone wants the same experience. The announcement comes at a pivotal moment as AI-generated content floods social media platforms worldwide. With more than 1.3 billion videos on TikTok already labeled as AI-generated, the platform recognized that users needed more granular control over their content preferences. Whether you're fascinated by AI creativity or prefer authentic, human-made content, TikTok now lets you customize your feed accordingly. How to Access the New AI Content Controls You can access the new capability by going into your Settings, selecting "Content Preferences" and then clicking the "Manage Topics" option. The interface is straightforward and intuitive, designed for users of all technical skill levels. Once you're in the Manage Topics section, you'll see a slider specifically for AI-generated content alongside other content categories like Dance, Sports, Food & Drinks, and Travel. Simply move the slider left to see less AI content or right to see more. The change is rolling out in the coming weeks, so if you don't see it yet, check back soon. What the Slider Actually Does It's important to understand what this feature can and cannot do. The AI content slider influences the recommendation algorithm's frequency for showing AI-generated videos in your For You feed—it doesn't completely eliminate them. TikTok uses this preference as one of many signals to personalize your experience while still exposing you to diverse content types. Think of it as telling TikTok, "I'm less interested in this type of content" rather than "never show me this again." The algorithm will prioritize other content types but may still occasionally surface AI-generated videos if they're particularly relevant to your other interests or trending widely. The Invisible Watermark Revolution Perhaps even more significant than user controls is TikTok's introduction of invisible watermarking technology. This represents a major leap forward in the platform's ability to detect and label AI-generated content accurately. Why Traditional Watermarks Aren't Enough TikTok already requires people to label realistic AI-generated content and uses a cross-industry technology called Content Credentials from C2PA, which embeds metadata into content that lets it and other platforms know when something is AI-generated. These Content Credentials work by attaching metadata to content files, creating a digital trail of how the content was created. However, there's a critical weakness in this system. When content is downloaded, edited in other apps, or reshared across platforms, these metadata signals can be stripped away or corrupted. A video created with AI on one platform might lose its identifying markers by the time it reaches TikTok, making it appear as authentic human-made content. How Invisible Watermarks Work With the new "invisible watermarks," TikTok will add another layer of safeguards by using a watermark that only it can read, making it harder for others to remove it. Unlike traditional visible watermarks that overlay text or logos on content, or metadata that can be stripped, invisible watermarks are embedded directly into the pixels of the video itself. These imperceptible patterns survive re-encoding, compression, screen recording, and most editing operations. While sophisticated tools might eventually be able to remove them, doing so would require significant effort and technical knowledge—creating a strong deterrent against misrepresenting AI content as human-created. Where TikTok Will Apply Invisible Watermarks TikTok will start adding invisible watermarks to AI-generated content made with TikTok tools like AI Editor Pro and is also adding them to content uploaded with C2PA's Content Credentials. This two-pronged approach means: Native AI Content: Any video created using TikTok's own AI tools automatically receives an invisible watermark at creation Imported AI Content: Videos uploaded from other platforms that include C2PA metadata will have invisible watermarks added upon upload This comprehensive coverage should dramatically improve TikTok's ability to maintain accurate labeling even as content circulates and gets reshared. Understanding AI-Generated Content on TikTok Before diving deeper into controls and detection, it's worth clarifying what TikTok actually considers AI-generated content. What Qualifies as AI-Generated AI-generated content includes images, video and/or audio that is generated or modified by artificial intelligence, such as artificial visuals, videos, or sounds that may portray realistic human likenesses or depictions created in a particular artistic style. Specific examples include: Videos featuring real people whose image, voice, or words have been altered by AI Scenes or events from the real world that have been modified by AI Entirely AI-generated videos or images of real or fictional people, places, and events Realistic deepfakes or synthetic media What Doesn't Require Labeling Not everything touched by AI needs to be labeled. TikTok's own effects and filters, even those using AI technology, are automatically labeled by the platform when applied. Simple enhancements like color correction, basic filters, or standard video editing tools don't trigger labeling requirements. The key distinction is whether the AI significantly alters reality in ways that could mislead viewers. A stylized filter that makes you look like a cartoon? That's obviously artificial. A deepfake making it appear you said something you never said? That absolutely requires disclosure. The Creator's Responsibility TikTok places significant responsibility on content creators to properly label their AI-generated content. Understanding these obligations is crucial for anyone creating content on the platform. Mandatory Labeling Requirements We require creators to label all AI-generated content that contains realistic images, audio, and video. Creators can disclose content as AI-generated by adding text, a hashtag sticker like #AIgenerated, or context in the post's description. When uploading content, creators see a toggle labeled "AI-generated content" in the More options menu. Activating this toggle adds a label that reads "creator labeled as AI-generated" below the video caption. This label cannot be removed after posting, ensuring transparency throughout the content's lifecycle. Consequences of Mislabeling The platform takes labeling violations seriously. Misleadingly labeling unaltered content with the AI-generated label violates TikTok's Terms of Service and may result in content removal or account penalties. Conversely, failing to label realistic AI-generated content can also lead to enforcement actions. TikTok's moderation team, which includes over 40,000 human moderators specifically trained to identify synthetic media violations, actively reviews content for proper labeling. The combination of automated detection systems and human oversight creates a robust enforcement framework. Why This Matters: The Deepfake Problem The introduction of these features isn't happening in a vacuum. The proliferation of AI-generated content, particularly deepfakes, has created serious challenges for online platforms and society at large. The Scale of the Issue As AI tools become more accessible and sophisticated, the volume of synthetic media has exploded. While much AI-generated content is harmless creative expression—art, entertainment, educational demonstrations—there's also significant potential for misuse. Deepfakes depicting real people in false scenarios, AI-generated misinformation about current events, fake endorsements by public figures, and synthetic media designed to manipulate political discourse all pose genuine threats. These concerns intensify during election cycles, public health crises, and other moments when accurate information is critical. TikTok's Multi-Layered Approach TikTok has implemented several safeguards beyond just user controls and watermarking: Prohibited Content: The platform completely bans certain types of AI content regardless of labeling, including deepfakes of private individuals without consent, AI-generated content depicting minors in any sexualized context, fake authoritative sources or crisis events, and content that harmfully misleads or impersonates others. Monetization Restrictions: AI-generated content is prohibited from TikTok's Creator Fund programs, removing financial incentives for mass-producing synthetic media. Educational Initiatives: TikTok announced it's launching a $2 million AI literacy fund aimed at experts, like the nonprofit Girls Who Code, to create content that teaches people about AI literacy and safety. Comparing TikTok to Other Platforms TikTok isn't alone in grappling with AI-generated content. How do its approaches compare to other major platforms? Pinterest's Similar Struggles The update follows a similar move from Pinterest, which has faced complaints about AI slop drowning out authentic images on the platform. Pinterest users have been particularly vocal about low-quality AI-generated images flooding search results and recommendations, diluting the platform's value for discovering authentic inspiration. Pinterest has implemented its own filtering options, though reports suggest the AI content problem remains significant. The challenge highlights how quickly AI content can overwhelm organic human creativity when platforms don't proactively manage the balance. Meta and OpenAI's Different Directions Interestingly, while TikTok moves to give users more control over AI content exposure, other companies are embracing AI-only feeds. Meta and OpenAI are experimenting with recommendation systems powered entirely by AI, both in terms of what content is shown and increasingly, the content itself being AI-generated. This divergence in approach reflects differing philosophies about AI's role in social media. Will users prefer curated human experiences or algorithmically optimized AI-generated content? TikTok's choice to let users decide for themselves may prove the wisest strategy. The Technical Challenges of Detection Even with invisible watermarks and C2PA metadata, detecting AI-generated content remains technically challenging. Understanding these limitations helps set realistic expectations. When Detection Systems Fail Currently, TikTok relies on a widely-used watermarking system called Content Credentials that adds metadata to AI-generated content, but these signals can get harder to detect when images or videos are edited in other apps or copied and reshared to other sites. Content that passes through multiple platforms and editing tools can lose its identifying markers. A video created with AI, posted to Instagram, screen-recorded, edited in a third-party app, and then uploaded to TikTok might appear entirely clean of any AI indicators by the time it arrives. The Arms Race History suggests that any detection system will face determined attempts at circumvention. Just as watermark removal tools proliferate across the internet, we can expect similar tools targeting invisible watermarks once their implementation details become better understood. TikTok's invisible watermarking represents a significant technical advance, but it's not a permanent solution. The platform will need continuous innovation to stay ahead of those seeking to pass off AI content as authentic human creation. How Creators Should Respond If you're creating content on TikTok, these changes have important implications for your strategy and workflow. Transparency Builds Trust Rather than viewing AI labeling as a stigma, embrace it as an opportunity for transparency. Audiences increasingly value authenticity and honesty. Being upfront about your use of AI tools can actually strengthen your connection with viewers rather than diminishing it. Many successful creators now incorporate their AI workflow into their content narrative, showing behind-the-scenes looks at how they use tools like AI Editor Pro or explaining which elements are AI-assisted versus traditional creation. Stay Updated on Policy Changes TikTok's AI content policies are evolving rapidly as the technology and its implications develop. What's allowed today might be restricted tomorrow, and vice versa. Regularly review TikTok's Community Guidelines and stay informed about policy updates. The platform's Creator Academy offers resources specifically focused on AI content guidelines, proper labeling techniques, and best practices for incorporating AI tools while maintaining authenticity. Experiment Within Boundaries AI tools open incredible creative possibilities. Don't shy away from experimentation—just do it responsibly. Test AI effects, try synthetic voices, generate background elements, and explore what's possible. The key is ensuring you're always transparent about what's AI-generated and what's not. Content that clearly showcases AI as part of the creative process rather than trying to hide it often performs better anyway. Viewers are curious about these tools and interested in seeing creative applications. The User Perspective: Making Your Choice For everyday TikTok users, the new controls offer an opportunity to shape your experience more deliberately. When to Dial Down AI Content You might want to reduce AI-generated content in your feed if: You value authenticity and human creativity over algorithmic generation You find AI content repetitive or lacking emotional depth You're concerned about misinformation or deceptive synthetic media You want to support human creators more directly You simply prefer "real" content over synthetic alternatives Sliding the AI content control toward "less" signals to TikTok that you want more human-created videos prioritized in your recommendations. When to Embrace AI Content Conversely, you might increase AI content if: You're fascinated by AI capabilities and want to see what's possible You enjoy the aesthetic qualities of AI-generated art and video You're researching AI tools or studying synthetic media You appreciate the experimental nature of AI creativity You want exposure to cutting-edge content creation techniques There's no right or wrong answer—it's entirely about personal preference. TikTok's approach respects that different users want different experiences. Finding Your Balance Most users will likely land somewhere in the middle, not maximizing or minimizing AI content but finding a comfortable balance. You can always adjust the slider as your preferences evolve or as the quality and nature of AI content on the platform changes. Pay attention to how different settings affect your For You feed and adjust accordingly. If you start seeing too many AI videos that don't interest you, slide it down. If you're curious about more, slide it up. The system learns from these inputs to better serve your preferences over time. The Broader Implications for Social Media TikTok's moves signal important trends that will likely influence other platforms and shape the future of social media. User Agency as a Priority By giving users direct control over AI content exposure, TikTok acknowledges that platform-knows-best algorithmic control isn't always optimal. This represents a shift toward user agency—letting people actively shape their experiences rather than passively receiving whatever the algorithm serves. We may see similar controls proliferate across other platforms as users demand more say in what they see. The era of purely algorithmic feeds might give way to hybrid approaches that blend algorithmic efficiency with user-specified preferences. The Authentication Challenge As AI-generated content becomes increasingly realistic and prevalent, authentication mechanisms will become critical infrastructure for digital platforms. Invisible watermarking, Content Credentials, and other provenance technologies may become standard features across the internet. We're moving toward a future where the default assumption may be that content could be synthetic unless proven otherwise. This inversion of trust has profound implications for everything from journalism to personal communications. The Creative Economy Shift These changes will reshape the creator economy. Content creators will need to navigate new questions about authenticity, transparency, and the role of AI tools in their creative process. Those who thoughtfully integrate AI while maintaining genuine connection with audiences will likely thrive. Conversely, creators who try to pass off AI-generated content as authentic human work may face growing skepticism and potential platform penalties. Transparency becomes not just an ethical imperative but a practical necessity. Looking Ahead: What Comes Next TikTok's rollout of AI content controls and invisible watermarking is just beginning. What developments can we anticipate? More Granular Controls The current implementation offers a simple more/less slider for AI content broadly. Future iterations might allow more nuanced controls—perhaps separating AI-generated humans from AI-generated art, or distinguishing between different AI styles and tools. Users might eventually customize their preferences for specific types of AI content while maintaining openness to others, creating highly personalized experiences that reflect individual values and interests. Improved Detection Accuracy As invisible watermarking technology matures and more platforms adopt Content Credentials, the accuracy of AI content detection should improve significantly. The current system catches much but not all AI content—future systems may approach near-perfect detection rates. However, this will require industry-wide cooperation and standardization. If some platforms refuse to implement these technologies, they become havens for unlabeled AI content that then spreads elsewhere. Educational Evolution The $2 million AI literacy fund represents TikTok's recognition that technology alone isn't sufficient. Users need education about AI capabilities, limitations, and implications to make informed choices about the content they create and consume. Expect more educational content, media literacy initiatives, and tools that help users develop critical thinking skills for evaluating digital content in an age of increasingly sophisticated synthetic media. Practical Tips for Navigating the New TikTok Ready to make the most of these new features? Here's your action plan: 1. Review Your Settings: Navigate to Settings > Content Preferences > Manage Topics and locate the AI-generated content slider. Experiment with different positions to see how they affect your feed. 2. Evaluate Your Feed: Spend a few days observing what appears in your For You feed with different settings. Notice which types of AI content you actually enjoy versus those that feel inauthentic or low-quality. 3. Report Mislabeled Content: If you encounter AI-generated content that isn't properly labeled, report it. Press and hold the video, tap Report, select Misinformation, then Deepfakes, synthetic media, and manipulated media. Your reports help improve system accuracy. 4. Label Your Own Content: If you create content using AI tools, always use proper labels. Toggle the AI-generated content setting when posting to maintain transparency and comply with platform policies. 5. Stay Informed: Follow TikTok's official announcements and creator resources to stay updated on policy changes and new features related to AI content. 6. Engage Critically: Develop habits of critical evaluation when viewing content. Ask yourself whether something might be AI-generated, look for telltale signs, and check for labels before sharing content that makes surprising or controversial claims. The Bottom Line TikTok's introduction of user-controlled AI content filtering and invisible watermarking represents a thoughtful approach to one of social media's most pressing challenges. Rather than dictating from above what users should experience, the platform empowers individuals to make their own choices while providing technological safeguards against deception. These features won't solve all problems related to AI-generated content. Determined actors will still find ways to spread unlabeled synthetic media, and detection systems will never be perfect. But they represent meaningful progress toward a social media ecosystem where users maintain agency over their experiences and transparency about content origins becomes the norm rather than the exception. As AI capabilities continue advancing at breathtaking speed, platforms like TikTok face ongoing challenges in balancing innovation with integrity, creative expression with authenticity, and algorithmic efficiency with user control. The approaches being pioneered today will shape digital media for years to come. Whether you choose to embrace AI content, minimize it, or find a middle ground, the important thing is that TikTok is now giving you that choice. In an era of rapidly evolving technology, user agency matters more than ever—and these new tools help ensure you can shape your own digital experience according to your values and preferences.