Skip to content
Learn Motion
ExploreHow it worksMembership
Log in
Learn Motion

Spotting AI-Generated Images and Videos: A Practical Checklist

1Why Detection Is a Judgment, Not a Test2Visual Clues in AI-Generated Images3Visual Clues in AI-Generated Video4Context and Content Clues5Metadata and Provenance Signals6A Practical Checklist and Common Traps
Context and Content Clues

Reading the source, not the follower count

3 / 3
The repost chain is the trap this page is built to catch. Four accounts, four captions, one file — it looks like four sources agreeing, but it is one source moving. The compression artifacts and the original upload time give it away. Real corroboration needs a different origin: a local outlet with its own footage, or a bystander filming from another angle. Until you find that, the apparent agreement adds nothing, and follower count does not change it.
0:00 / 0:00

Judge a source by history, incentives, and independent corroboration. Follower count and production polish are weak signals that are easy to fake or inherit.

What actually carries weight

Strong signals

  • A track record of verifiable posts
  • An independent source confirming the same event
  • A clear, checkable chain back to an original

Weak signals

  • High follower count
  • Professional-looking editing
  • A second account reposting the same file

The independence rule

Two accounts sharing one file are one source wearing two names. Independent confirmation requires a different origin — a different photographer, agency, or vantage point — not a second copy of the same media.

A repost chain that looks like confirmation

A dramatic clip appears on a large account, then on three smaller ones within an hour, each with slightly different wording. It looks corroborated. But all four carry the same file, the same compression artifacts, and the same original upload time. Nothing independent has confirmed the event; the appearance of confirmation is just the same file moving. Finding one separate source — a local news outlet with its own footage, or a bystander with a different angle — is what would actually change the judgment.

Previous3 / 3Next

Learn Motion

Generate a course. Learn it properly.

Operated by Wuhan Daoyin Technology Co., Ltd.

Contact: [email protected]
Privacy PolicyTerms of Service

© 2026 Learn Motion