When an illustration moves from where its maker put it to somewhere else, something usually falls off along the way: the name. It is rarely malice. Metadata is the first casualty of any copy operation, because a copy is made by whoever wants the picture and the picture is the only part they wanted. The consequence is the same either way.

Credit is how the money moves

Most independent illustrators are not paid per view. They are paid by commissions, subscriptions, print sales and direct support, and every one of those depends on somebody being able to find them after seeing the work. The image is an advertisement for a service; without the name attached, the advertisement runs and the business never hears about it.

That is why artists react to uncredited re-hosting more strongly than outsiders expect. It is not that a copy exists. It is that the copy is doing the job of the original while routing none of the return. A credited repost, by contrast, is often welcomed, because it does exactly what the artist wanted the image to do.

Why metadata disappears by default

Right-click and save produces a file with no author. Screenshotting removes even the embedded fields. Many upload pipelines strip EXIF deliberately, for privacy reasons that are entirely reasonable in other contexts. Re-encoding to a smaller format drops what is left. At no point does anybody decide to remove the name; the tools simply do not carry it.

Caption-based credit is even more fragile, because a caption lives on the page rather than in the file. The moment somebody saves the image and posts it elsewhere, the caption stays behind. Chains of three or four reposts are common, and by the third the origin is usually unrecoverable without a reverse image search.

What actually survives

A signature drawn into the image survives everything except a crop, and crops are common but not universal. A watermark across the middle survives crops too, at the cost of the picture. Many artists compromise: a small signature in a corner plus a second faint one somewhere less obvious, so a crop removes one and leaves the other.

What a re-host could do

Restoring credit is technically easy. Reverse image search finds most origins, an artist field on an upload form takes a minute to add, and linking out costs nothing. It is not done because it produces no traffic, and a site funded by traffic will not spend effort on something that sends people away.

Sites that do maintain credit tend to be built around preservation or community rather than advertising, and artists treat them very differently as a result. The distinction between a collection that keeps things and a collection that gathers things shows up in exactly this behaviour, which is also why takedown handling correlates so strongly with credit handling.

There is also a slower mechanism at work. Once a name has been detached from an image for long enough, the image starts to feel ownerless, and people who would never knowingly strip a credit will reuse it without one because there is nothing visible to preserve. Attribution decays through indifference rather than through theft, which is why the early copies matter far more than the late ones.

Keeping your name attached to your work

Practical steps for an illustrator: sign the image itself, keep a public gallery you control, use a consistent handle across platforms so searches converge, and post the highest-quality version where you want people to arrive rather than where they happen to be. None of these stop copying. All of them raise the chance that a copy leads somewhere.

This site publishes craft notes rather than files, and the reasoning is set out in our editorial principles. If you want to reach us about attribution, use the letters page, and creation itself stays on explore page 3 where accounts and rights controls live. The related economics are covered in the piece on where the money comes from.