# Can a watermark survive a screenshot? Testing the attacks

> The screenshot defeats most image protection because it re-captures only what is on screen. Here is why a pixel-level watermark survives it when metadata and blockers do not.

Source: https://tilkie.app/learn/watermark-survive-screenshot/
Published: 2026-06-26
Publisher: Tilkie (INFOWISE LIMITED)

---

The screenshot is the simplest attack and the one that quietly defeats most image protection. It needs no tools and no skill, and it strips away several kinds of "protection" in one motion. So it is the right test for any watermark: does it survive being screenshotted?

## Why the screenshot is so effective

A screenshot captures only what is drawn on the screen. That single fact has big consequences:

- It produces a brand-new image file, so anything stored outside the visible pixels is gone. Metadata, hidden file layers, and overlay tricks do not make it into the capture.
- It cannot be blocked by a website, because the operating system takes the screenshot, not the browser. (More on that in [disable right-click and other lies](/learn/disable-right-click-myth/).)
- It re-encodes the image, usually as a fresh PNG or JPEG, which is its own round of compression.

So the screenshot is really several attacks at once: it discards non-pixel data, dodges every browser-side blocker, and recompresses the result.

## What each approach does under a screenshot

- **Metadata and EXIF copyright fields:** gone immediately. The screenshot never includes them, the same way uploads strip them, which is why [metadata is not protection](/learn/metadata-exif-not-protection/).
- **Right-click and copy blockers:** irrelevant. The capture happens at the operating-system level, where the script has no reach.
- **A visible logo:** survives the screenshot itself, but only until the thief crops it out of the captured image, which takes seconds.
- **A resilient invisible watermark:** survives, because it lives in the visible pixels themselves. If the product is visible enough to be worth screenshotting, the mark is in the part that got captured.

That last point is the key. A pixel-level watermark is not hidden in a layer the screenshot drops. It is embedded in the image you can see, so the act of capturing the image also captures the mark.

## Surviving the capture is not automatic

A screenshot still re-encodes and sometimes rescales the image, which would break a fragile watermark. This is where resilience matters: spreading the code across the whole frame and adding error correction so a recompressed, resized capture still reads. We cover how that works in [why a good watermark survives cropping, resizing, and re-encoding](/learn/surviving-crop-resize-reencode/). A watermark built and tested for these conditions treats a screenshot as just another manipulation it was designed to handle.

## The takeaway

You cannot stop someone taking a screenshot, and nothing honestly can. The realistic goal is for the screenshot to carry your mark anyway, so that even a captured, cropped, reposted copy can still be proven yours. That is what an [invisible watermark](/learn/what-is-invisible-watermarking/) is built to do, and it is why Tilkie protects the image itself rather than trying to lock the page it sits on.