Who is infringing copyright when hotlinking is involved?

Question

When a page at "Pirate Site" embeds images from "Example Site" by hotlinking, we have the following situation:

• Pirate Site isn't actually copying, storing, or manipulating anything. The pages only contain code with instructions that tell the browser to go fetch the images on Example Site. Example of code: <img src="https://www.example.com/copyrighted-image.jpg">
• The visitor's browser automatically downloads the images from Example Site and displays them on Pirate Site. The visitor's browser is what is actually doing the work: going to get the image, downloading it, storing a local copy temporarily, and displaying it.
• The visitor will do all that automatically without even realizing it or meaning it, just by visiting Pirate Site, even if they landed on that site by mistake.

We apparently have a paradox where the real "pirate" (the owner of Pirate Site) isn't even touching any copyrighted data in any way, but on the other hand the one who is actually processing the copyrighted material (the visitor) cannot avoid it. To avoid infringement, the visitor would have to open the source code before the browser executes it, check all the code, check who the copyright holders are, and make sure no copyrighted material will be downloaded when the web page is opened. So the pirate isn't touching the data, and the visitor technically can't avoid breaking the law.

Does this paradox have a solution? Who is infringing the copyright here?

Answer 1

According to EU case law, everything in your scenario is legal except if Example Site is hosting the image without authorization and Pirate Site is a for-profit site, then Pirate Site is presumed to be violating Article 3 of the Copyright Directive on communication to the public (in this scenario, Example Site is also trivially violating Article 2 on the right to reproduction).

---

In Meltwater, Case C-360/13, the court ruled that browser cache and on-screen copies fell under the temporary reproduction exception, Article 5(1) of the Copyright Directive. This means that the visitor is not infringing copyright (IPKat reference).

In BestWater, Case C-348/13, the court ruled that embedding content was itself not a communication to the public when that content was hosted with rightsholder authorization, and so did not violate Article 3. This means that Pirate Site is not infringing on communication to the public rights (it is also not creating a copy itself, so is not breaking Article 2) (IPKat reference).

When content is not hosted with authorization, the situation is quite a bit more nuanced. GS Media, Case C-160/15, is the controlling case. Here, the court ruled that if a link (note it doesn't even have to be embedded/hotlinked) is posted by a for-profit site, that site is expected to have done its due diligence to ensure the linked content is hosted legally. Therefore, it is presumed to be violating Article 3, i.e., the burden of proof is on the link posting site to demonstrate that it had done its due diligence in verifying the legality of the linked content. So in this scenario, Pirate Site is presumed to be infringing on communication to the public rights (IPKat reference - WARNING: slightly NSFW image here, Playboy was one of the parties to the case).

Answer 2

At least some courts have held that a link, in which the address of copyrighted content is provided, but the accused infringer never hosted or directly copied the content, in not an infringement at all, whether or not it is framed.

If the views of those courts prevail, no one in the scenario described in the question is infringing any copyright.

The law is not yet finally decided on whether such links violate copyright or not. My personal view is that they do not, and should not, but the courts may decide otherwise.