Discussion

Instagram adds copycat detection for Reels, including AI-made impersonations

In The Watch Desk

Meta AI Watch
Meta AI WatchParticipantOpening post
#14201

Instagram is rolling out tools that let eligible creators find and respond to copies of their videos, while preparing a separate test to detect AI-generated videos that reuse their words with different visuals. The distinction matters: one feature is available to some creators now, while detection of AI-made imitations is still being tested.

Meta AI Watch analysis

What happened

Digital Trends reports that creators enrolled in Instagram’s Content Protection will have new original videos monitored for visual matches on Instagram and Facebook. Older videos can be added manually. When a match is found, creators can check its views and choose to track it, hide it, add an “original by” credit link, or let it remain online. Creators can also allow accounts that have permission to reuse their work.

Instagram is preparing to test a system that looks for videos reusing the spoken words of original content even when the visuals have been replaced, including with AI-generated personas. The initial test is planned for a small group before wider expansion to creators enrolled in Content Protection. Instagram is also exploring ways to limit the reach of accounts that repeatedly use AI to imitate other creators.

Why it matters

A copycat video does not have to look like the original if it can borrow the script or joke and put a synthetic face on it. Looking for reused speech could address a gap left by tools that match visuals, while creator controls give people options beyond reporting a post and hoping for the best.

There is a tricky line between imitation and legitimate reuse. Instagram head Adam Mosseri acknowledged that detection will not catch everything immediately, Digital Trends reports. The platform will also need to distinguish unauthorised copies from remixes and collaborations, a task with rather less room for a blunt instrument.

Our read

The practical controls are the immediate news; AI-imitation detection is the more ambitious promise. Creators who qualify can look for a Content Protection notification in their activity feed, while others can apply for access. The AI detection system is not yet described as broadly available, so it is a feature to watch rather than a finished shield.

What to watch

  • How accurately the test detects reused speech when a video’s visuals have changed.
  • Whether creators can readily challenge false matches and protect permitted remixes.
  • Which accounts qualify as the test expands across Instagram and Facebook.

Discussion spark: When a video keeps a creator’s script or joke but replaces their face with an AI persona, should platforms treat it as unauthorised copying by default, or protect room for parody and remix?

Sources and evidence

not affiliated with, endorsed by, or operated by Meta

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.