in News
at October 09, 2026

Common Mistakes When Sharing Breaking News on Social Media

A video of smoke pouring from a building lands in your feed. No location tag, no date, four hundred shares already. You have a caption half-written. This is the moment where most bad sharing happens — not out of malice, but out of speed. The mistakes below are the ones that show up again and again: the photograph from another year, the clip with no address, the casualty figure that arrived three hours too early. None of them need a journalism degree to avoid. They need a pause, a search box, and the willingness to type the words "not yet confirmed".

The clip with no address

Footage travels without paperwork. A fire, a crash, a crowd running down a street — the images are generic enough to belong to a hundred cities, and often do. The clip is real. The story attached to it is the part that goes wrong.

Before you share, try to answer three questions: who filmed this, when, and where were they standing? If you can answer only one of the three, you are not sharing news. You are sharing a rumour with good production values.

Open the original post rather than a reshare. A repost that credits "social media" credits nobody, and the trail ends there. Look at the account that uploaded it: what else does it post, how old is it, does it have any connection to the place in the video? Then look inside the frame. Street signs, shopfronts, the language on a police jacket, the weather, the trees, the number plates. Pull a still and run it through a reverse image search. Check whether a local newsroom or the fire service has said anything at all.

One more habit worth building: name your source in the post itself. "Footage posted by a resident on X, location unverified" is a sentence that protects you and informs your readers.

The photograph that is older than the story

Old images resurface at every major event. They are dramatic, they are indexed, and search engines do not care what year it is. A picture of a flooded street from a previous decade can be circulating within an hour of a new storm, shared by people who genuinely believe they are helping. The second wave is worse than the first, because by then it has been screenshotted and stripped of everything that would let you check it.

A quick routine catches most of them:

Search the image, not the story. Find the earliest appearance, not the most popular one.
Check the season. Snow, foliage, and clothing give away the month even when the caption does not.
Read the small details: licence plates, shop names, uniforms, the model of a phone held up in the crowd.
Look at the uploader's history. Accounts that post the same disaster footage in five countries are telling you something.
Treat every screenshot as evidence of nothing. It has lost its date, its author, and its replies.

If a photo is genuinely from today, saying so takes ten seconds. If you cannot establish that, saying nothing takes less.

Context is more than a caption

A clip that begins mid-argument is not a clip of an argument beginning. A quote published without the question before it can invert its meaning. A long lens compresses a crowd of two hundred into something that looks like a mob filling a square. None of this requires editing software. It only requires a crop and a decision to leave the rest out.

So describe the frame, and then describe the edge of the frame. "This video shows a scuffle outside the station. What happened before it started is not visible here." That sentence is not a weakness. It is the most useful thing you will write all evening.

Be just as careful with cause. Avoid the word "because" unless someone with authority has used it, or unless the video actually shows the cause. Most of the time it does not.

The numbers, the names, and the cause

Three things are reliably wrong in the first hours: casualty figures, the identity of victims, and what caused the incident. Numbers move as responders reach the scene and hospitals report in. Causes are investigated for weeks. And names should come from officials, after families have been told — never from a scanner feed, a comment thread, or a relative's public profile.

Attribute everything. "According to the fire service" is a complete thought in a post. If you cannot name where a detail came from, you do not have a source, you have an impression.

One caution worth stating plainly: if a post could identify a private individual or imply that someone has committed a crime, slow down. Defamation and privacy rules vary by country and by platform, and the time to take advice is before you publish, not after a letter arrives. If you are unsure whether a claim about a named person is safe to repeat, that uncertainty is your answer.

During a live incident, your post is part of the scene

Do not publish the location of responders, staging areas, or the movements of police units. Do not livestream tactical positions. Do not translate an unconfirmed report of a second incident into a warning, because a false alarm in a tense hour sends panic through neighbourhoods and ties up emergency lines that somebody else needs.

Share the official account's post instead of paraphrasing it. If you must summarise, quote exactly and say where it came from. And resist the urge to add the detail you heard from a friend of a friend; that detail is how a small error becomes a large one.

When you get it wrong, say so in the same place

Corrections are the part most people skip. If a post has spread, the correction has to travel at least as far. Reply to your own post rather than quietly deleting it — a deletion looks like tidying, and the people who shared it will never learn anything. Edit with a clear note at the top, in plain words: what was wrong, what is right, and when you changed it.

Do not reuse the confident tone for the correction that you used for the error. And do not delete a mistaken post and republish the corrected version as though it were new. That erases the record and leaves everyone who shared the first version holding something you have quietly abandoned.