Skip to main content
GetProofAnchor logo
GetProofAnchor
Verifiable web evidence
Dashboard
← Back to all articles

How to tell if a screenshot is fake, and how to prove yours is real

The red flags that catch careless fakes, the one test that works better than any of them, and how to capture online content so nobody has to trust a picture.

Screenshots Verification Digital evidence eIDAS

The honest answer: the image alone rarely tells you

If you came here hoping for a single trick that exposes a fake screenshot, here is the uncomfortable truth: in most cases you cannot prove a screenshot is fake, or real, just by looking at the image. A screenshot is a picture of what a screen displayed, and whoever controls the screen controls the picture.

The most convincing fakes are not edited in an image editor at all. Every modern browser lets the user change the text of a loaded page before taking the screenshot, through the developer tools or a single setting such as document.designMode. The browser then renders the altered text with the real fonts, the real layout and the real logo. There are no cloned pixels and no compression seams to find, because nothing was pasted in. We demonstrated this in our own experiment, linked at the end of this guide.

Red flags that are still worth checking

Image checks cannot prove authenticity, but they do catch lazy fakes. Look for:

  • Fonts, sizes or letter spacing that differ between one line and the rest, or text that does not line up with its neighbours.
  • A status bar that does not fit the story: the clock, date, battery level, carrier or notification icons contradict when and where the screenshot was supposedly taken.
  • Interface details from the wrong app version, operating system or theme, such as an outdated button style or a light element inside a dark-mode screen.
  • Names, profile pictures or handles that do not match the account as it looks today, or as it looked in older captures.
  • Tight cropping that removes the address bar, the date, the sender or the rest of the conversation.
  • Blurry or blocky patches around a single word or number, which can point to a pasted or retouched area.

Metadata is a weak signal in both directions. Messaging apps and social networks often strip it, so missing metadata proves nothing, and metadata that is present can be edited. Treat every red flag as a reason to check the source, not as a verdict.

The strongest test: go back to the source

Because the image is the weakest link, the checks that actually work happen outside it:

  • Ask for the link. A real post, message or web page has an address you can open yourself and compare.
  • Check the platform's own record: the post itself, its edit history, the message in your own app or a data export of the conversation.
  • Look for independent copies: other people's captures, quotes in the press or public web archives that saved the page at the time.
  • Compare the context: does the wording, tone and timing match what the account normally publishes?

If the content is already gone and nobody captured the original properly, you may never know for certain. That is why the moment of capture matters more than any later analysis.

What forensic examiners can and cannot do

Image forensics can sometimes detect splicing, resampling or inconsistent compression in an edited image file. It cannot tell whether the text on the screen was true when the screen was rendered. A screenshot of a page that was altered in the browser is, technically, a genuine screenshot of an altered page.

So examiners, opponents and judges tend to ask different questions: who captured the content, when, from which address, with what method, and has the file changed since? A plain screenshot answers none of them. Provenance standards such as C2PA content credentials help for images from cameras and editing tools that support them, but a typical phone or desktop screenshot carries no such credential.

How to make your own capture provable

If you are the one who needs to rely on online content, do not stop at a screenshot. A capture that is hard to dismiss has four properties:

  • It records the source, not just a picture: the address, the page code, the network traffic and the server certificate, not only pixels.
  • Every file has a cryptographic fingerprint (SHA-256), so any later change is detectable.
  • The fingerprints are sealed with a qualified electronic timestamp. Under Article 41(2) of the eIDAS Regulation a qualified timestamp enjoys the presumption of the accuracy of its date and time and of the integrity of the data it is bound to, and under Article 41(3) it is recognised in all EU member states.
  • Anyone can verify it later without having to trust the person, or the company, that made it.

Know the limits: a timestamp proves that a specific set of files existed unchanged at a specific time. It does not prove that what the page said was true. That question stays with whoever you present the evidence to.

Three ways to do it with GetProofAnchor

GetProofAnchor turns a web page into an evidence package instead of a picture. Pick the route that fits your situation:

  • Web app: enter the address and our servers capture the page (full-page screenshot, HTML, MHTML snapshot, video, network log, TLS certificate chain, DNS) and seal it with a qualified eIDAS timestamp from SK ID Solutions, an EU qualified trust service provider. The starter pack costs 9 € + VAT (10.89 € incl. VAT) for 3 proofs.
  • Browser extension: for content you only see after logging in, the page is captured in your own browser and sealed when our server receives it. The timestamp then proves integrity and the time of receipt.
  • Assisted Capture: send us the links and an operator captures them for you, no account needed. Single from 49 € + VAT (59.29 € incl. VAT) within 90 minutes; Rush for up to 5 links with a 60-minute guarantee for 249 € + VAT (301.29 € incl. VAT).
  • Verification: every package can be checked on our verify page or offline with the open-source gpa-verify tool, which checks the fingerprints and the qualified timestamp without relying on our servers.

Whichever route you choose, the evidence no longer depends on anyone believing an image.

Checklist: someone sent you a screenshot

  • Do not forward it or react publicly before you have checked it.
  • Ask for the link or the original message and open it yourself.
  • Look at the status bar, fonts, names and cropping for obvious inconsistencies.
  • Search for independent copies or archived versions from the same time.
  • If the content is real and matters to you, capture the source page properly now, before it is edited or deleted.
  • Keep the screenshot too, but as a pointer to the evidence, not as the evidence itself.

Bottom line

You will rarely prove a screenshot fake by staring at it. You can refuse to rely on screenshots alone, check the source, and make sure your own evidence answers the questions an opponent will ask: what, where, when, and unchanged since.

If a post, review, listing or message matters to you, capture the live page with a qualified timestamp while it still exists. It takes a few minutes, and it is the one step you cannot repeat once the content is gone.

FAQ

Can you tell if a screenshot is fake just by looking at it?
Sometimes you can spot a careless edit, such as mismatched fonts or a status bar that contradicts the story. A fake made by editing the page in the browser before taking the screenshot looks identical to a real one, so the reliable check is to compare it with the original source.
Does metadata prove that a screenshot is real?
No. Many apps strip metadata when an image is shared, and metadata can be edited. Its presence or absence is a hint at most.
Is a screenshot enough as evidence?
It is often submitted, but how much weight it gets depends on the case, and a contested screenshot without a source, date or integrity check is easy to challenge. A sealed capture of the source page is much harder to dismiss, although no method can guarantee how a court will assess it.
How can I prove that my screenshot is real?
After the fact, usually you cannot. What you can do is capture the original page again while it still exists, with its address, page code and a qualified timestamp, so the evidence no longer depends on the image.

Sources

Capture it before it disappears

Turn the page into evidence in 2 minutes

Paste the URL and get a sealed Evidence ZIP with an eIDAS qualified timestamp — much harder to challenge than a screenshot. No subscription needed.

3 proofs for €9 + VAT We capture it for you — from €49 + VAT

Verify an existing proof → · Guides for Airbnb, Vinted, Amazon & more →

Related guides

Don't rely on a screenshot. Capture the source.

Seal a web page with a qualified eIDAS timestamp in a few minutes, or send us the link and we capture it for you.

Create a proof Verify a proof

Starter pack 9 € + VAT (10.89 € incl. VAT) for 3 proofs, no subscription.