TikTok is testing a new AI likeness detection tool designed to help creators identify and report deepfakes and unauthorized AI-generated versions of themselves. The opt-in feature, currently in early testing with select US creators, allows users to scan for AI recreations on the platform and flag them for company review.
TikTok US spokesperson Zachary Kizer confirmed the initiative to The Verge, though the company provided limited details on how the detection system works or its accuracy rates. The tool represents TikTok's response to growing concerns about synthetic media abuse, where bad actors use AI to impersonate creators without consent, often for fraud, harassment, or sexual content.
This move mirrors YouTube's efforts to develop similar detection capabilities. Both platforms face pressure from creators and regulators to combat deepfakes and unauthorized digital replicas. The problem has grown with the proliferation of open-source AI models and accessible video generation tools that require minimal technical skill to operate.
The opt-in nature of TikTok's tool means creators must actively choose to use it rather than having automatic detection running across all accounts. This approach prioritizes user choice but risks leaving some creators unaware their likenesses are being misused.
Key questions remain unanswered. TikTok has not disclosed whether the tool uses on-device processing or server-side analysis, how it distinguishes AI content from legitimate parodies or tributes, or what happens after creators report detected likenesses. The company also hasn't clarified how quickly it removes flagged content or whether it will take legal action against repeat offenders.
The testing phase suggests TikTok is still refining the technology before wider rollout. Success depends on both detection accuracy and response speed. False positives could frustrate creators, while missed detections leave vulnerable users unprotected. TikTok's scale, with hundreds of millions of daily