Every few days, something explosive crosses your feed: a shocking quote, a dramatic photo, a "breaking" claim with fifty thousand reposts. Most people make a share-or-scroll decision in about three seconds, using exactly one input — whether the claim flatters what they already believe. This article is the upgrade: a verification routine that takes five minutes, requires no special tools, and catches the large majority of viral fakery. Consider it a home first-aid kit for information.
Step 1: Interrogate the account (60 seconds)
Before evaluating the claim, evaluate the claimant. Open the profile and check the trio we detail in our fake-accounts guide: join date, posting history, and handle spelling. A three-week-old account "breaking" international news is answering your question before you ask it. If the account claims to be an organization or public figure, remember that a blue checkmark is a subscription, not an identity check — confirm through the entity's own website instead.
Step 2: Reverse-search the image (90 seconds)
Miscaptioned real photos outnumber doctored ones enormously — the classic viral fake is a genuine image from a different year, country, or event. Right-click (or long-press) the image and search it with Google Lens or TinEye. If the "last night's protest" photo first appeared in 2014, case closed. For video, screenshot a distinctive frame and search that. While you're looking, apply the basic forensics: shadows pointing in different directions, mangled text on signs, hands and jewelry (still where AI generators stumble most), weather that doesn't match the claimed date and place.
Step 3: Hunt the primary source (90 seconds)
Viral quotes and statistics are usually orphans — stated with total confidence, attached to nothing. Search the exact quoted phrase in quotation marks. Three outcomes: it traces to a real speech, document, or study (read the surrounding context — selective quotation is the gentleman's fabrication); it traces only to other viral posts citing each other in a circle; or it traces to a satire site that lost its label somewhere around repost four hundred. Only the first outcome is evidence. For claims about what someone posted and then deleted, check the web archives — our archiving guide explains why a Wayback link settles what a screenshot can't.
Step 4: Triangulate (60 seconds)
If the event were real and significant, who else would know? Search the topic on a news aggregator and on the platform itself sorted by Latest (the search operators help here). Genuine major events produce coverage from multiple unrelated outlets within the hour, plus local sources, official statements, and eyewitness variety. A "massive story" that exists only as one viral post and its reposts is not being suppressed — it's being invented. Be especially wary during breaking news: the first hour of any real event is also the golden hour of fabrication, when hijacked trends fill the information vacuum with recycled footage and confident guesses.
A note on AI-generated media
Synthetic images and video have graduated from novelty to background condition, and the old tells (warped hands, garbled text) fade with every model generation. Two adjustments follow. First, shift weight from "does it look real" to provenance: where did this file first appear, who posted it, and does any credible outlet or original witness account anchor it? An image with no traceable origin is unverified regardless of how convincing it looks. Second, watch for the emerging content-credentials standards (C2PA and similar) that let cameras and publishers cryptographically sign media — adoption is uneven, but a signed image from a news organization and a context-free viral image now deserve very different default trust.
The emotional tripwire
The meta-skill above all four steps: outrage is a targeting signal. Fabricated content is engineered to produce exactly the surge you feel — fury, vindication, disgust — because that surge is what makes thumbs hit repost. Treat a strong emotional reaction to an unverified post the way you'd treat a stranger urgently demanding you wire money: as a reason for more scrutiny, not less. The posts most worth verifying are precisely the ones you most want to be true (or most want your opponents to be guilty of).
What to do with the verdict
- Confirmed: share it with the primary source attached — you'll inoculate your own audience against the sloppier versions.
- Debunked: resist the quote-repost dunk; engagement-ranked feeds amplify whatever you engage with, including lies you're correcting. If you must correct, do it without redistributing the original.
- Unclear: the hardest and most common outcome. Sit on it. "I couldn't verify this" is a complete reason not to share, and the story will either firm up within hours or quietly evaporate — which is itself the answer.
Five minutes feels expensive against a three-second habit. But you don't need to verify everything — only what you're about to believe, repeat, or repost. For everything else, the free option always works: just keep scrolling.