// AI NEWS · PLATFORM

TikTok Begins Testing an Opt-In Tool That Lets Creators Find and Report AI Deepfakes of Themselves

Following YouTube, TikTok is testing a likeness-detection tool with a small group of US creators. It scans for AI-generated videos using a creator's face, but requires ID verification through Jumio first — a real-time selfie plus a government ID.

KompozyTurn one idea into a week of content — across every platform, published for you.
Get Started →

2026-07-18 · by Moe Ameen

What happened

TikTok has begun testing a likeness-detection tool that lets creators find AI-generated videos using their face or likeness without permission, and report the posts or accounts behind them. TikTok US spokesperson Zachary Kizer confirmed the test to The Verge, saying it is limited to a small group of US-based creators. Coverage surfaced on July 17, 2026.

The tool is opt-in, but creators who want it must first verify their identity through Jumio, a third-party identity-verification provider. Verification includes a real-time selfie and a government-ID check. After a creator is verified, TikTok scans AI-generated content that may include their face and surfaces possible matches, which the creator can review and then report if they believe the content is impersonating them.

On privacy, Kizer said TikTok does not keep ID documents, and that facial data is used only to match a creator's likeness and identify possible unauthorized AI-generated content. That verification-first, opt-in design contrasts with default-in features like Meta's short-lived Muse Image tool, which drew backlash before Meta pulled it.

The move puts TikTok on the same path as YouTube, which launched its own likeness-detection tool in late 2025 and expanded access through 2026 — to eligible creators over 18, then to public figures and talent agencies. TikTok has not announced a general-availability date; as of mid-July 2026 this is a limited test, not a full launch.

Why it matters for creators

  • Creators finally get a first-party way to find deepfakes of themselves. Until now, spotting an unauthorized AI video using your face meant a follower flagging it or a manual search — an enrolled creator can review platform-surfaced matches directly.
  • Detection puts the wronged person in the loop with a report button, rather than waiting for a platform reviewer to notice. That is a meaningful shift in who gets to act on an impersonation.
  • The identity-verification gate is the trade-off. To use the tool you hand TikTok a real-time selfie and an ID via Jumio — TikTok says it doesn't retain the ID, but creators uneasy about biometric enrollment have a real choice to weigh.
  • It's a limited US test, not a shipped feature. Scope, eligibility, and the review-and-report flow can change before any wide rollout, so treat the specifics as a beta snapshot.
  • It sharpens the line between owned and borrowed identity. Detection tools exist to catch unauthorized use of a real person's face — content generated from an identity you actually own and can consent to is exactly what they are built to pass over.

How to act on this with Kompozy

There are two ways to act on this. The first is defensive and structural: the surest way to never land on the wrong side of a likeness scan is to only ever generate from a face you own. Kompozy's persona system is built that way — its Persona Shorts and avatar-video formats generate talking-head content from an AI Influencer persona you create and control, your own face and voice via a HeyGen avatar and Gemini face-lock, or a consistent synthetic character that is yours. Every asset is generated from a likeness you can consent to, which is precisely what detection tooling is designed to leave alone.

The second is to publish your take this week while the topic is hot. "TikTok is testing a tool to detect AI deepfakes of creators" is a query your audience is searching right now, and being early is the whole value. Drop your angle into Kompozy and one source fans into a captioned Clipped or Persona Short explaining the tool, a brand-exact Carousel breaking down the enroll-verify-scan-report flow, a Blog Article on what creators should do, Quote Graphics, and native Text Posts — held to one voice by your Persona Brief and scheduled across nine platforms plus blog and email from one review queue. TikTok gives verified creators a way to police their likeness; Kompozy is how you produce from an identity you own and turn a same-day news beat into a week of on-brand posts.

Quick takeaways

  • TikTok is testing an opt-in likeness-detection tool that lets verified creators find and report AI-generated videos using their face; coverage surfaced July 17, 2026.
  • It is limited to a small group of US creators and requires identity verification through Jumio — a real-time selfie plus a government-ID check — before use.
  • TikTok says it does not retain ID documents, and that facial data is used only to match a creator's likeness and flag possible unauthorized AI content.
  • The move follows YouTube, which shipped a likeness-detection tool in late 2025 and expanded it through 2026 to creators, public figures, and talent agencies.
  • The safest production posture is to generate from an identity you own — which is how Kompozy's face-locked persona system is built.

Frequently asked questions

What is TikTok's likeness detection tool?

It is an opt-in tool, in limited US testing as of July 2026, that scans AI-generated content for a creator's face and surfaces possible matches so the creator can review and report unauthorized deepfakes. It follows YouTube's similar tool. TikTok has not announced a general-availability date.

How do you use TikTok's likeness detection tool?

A creator opts in and first verifies their identity through Jumio, a third-party provider — a real-time selfie plus a government-ID check. Once verified, TikTok scans AI content that may include their face, shows possible matches, and lets the creator report posts or accounts they believe are impersonating them.

Does TikTok keep your ID or face data?

Per TikTok spokesperson Zachary Kizer, TikTok does not keep the ID documents used for verification, and facial data is used only to match a creator's likeness and identify possible unauthorized AI-generated content. Because it involves biometric enrollment, creators should weigh the trade-off before opting in.

Is this the same as YouTube's likeness detection?

It is the same category and points the same direction. YouTube launched its likeness-detection tool in late 2025 and expanded it through 2026 to eligible creators, public figures, and talent agencies, with enrollment in Studio via a QR code, ID, and selfie video. TikTok's test is earlier-stage and US-limited, but the enroll-verify-scan-report shape is broadly similar.

Related news

← All AI news · Get started →