It is tantamount to theft, and a growing concern since way back when.
Google and other SEs actually look for the
images folder on our server. That is where they like our images to be stored, at least the ones for which we want some potential of indexing.
When we hotlink, the crawler traces back to the source and credits them with the image, not our site. Too many of these tracing from our site and it is likely the the SE will move it into a supplemental index and we’ll likely never show up in SERPs. Hotlinking can kill our chances in the search game.
Depending upon how many images are in a particular page, we can give ourselves some advantage by using a CDN for images, and keep only source files on our host server. Typically, a client can only have two active ports on the server at any one time. But there is no such limit to how many servers we connect to, meaning if we have two servers, one for files and one for images, we can download them concurrently, two resources from one server, two from the other.