How to Spot a Fake Leak or Viral Claim Online

Short answer: Trace the claim back to its first post rather than the video repeating it. Then ask three questions: does this source have dated, verifiable hits from before announcements; is there a second independent report; and is the claim written to create urgency? A claim that fails all three is entertainment, not information.

Viral claims spread because verification is slow and reposting is instant. By the time a rumour is checked, it has usually been rewritten by a dozen accounts as established fact, each removing the hedging language of the last.

The five checks

1. Find the origin. Reverse-search the image and read the earliest post, not the summary.
2. Check the track record with dates. A screenshot of past accuracy proves nothing without timestamps preceding the announcement.
3. Look for independent confirmation from an outlet that names an editor and a correction policy.
4. Read the wording. 'Sources say' with no detail, plus a countdown, is a pattern.
5. Check whether the claim is falsifiable at all. Predictions vague enough to always be right are marketing.

Images and video need their own checks — see how to tell if a photo or video is AI-generated. For a worked example of applying this to a single story, read our breakdown of what is actually confirmed about GTA 6. The same discipline applied to research headlines is in how to check if a health claim is credible.