Originally Posted by
rdenney
The reason seems simple. A copyright isn’t attached to the work. It is what attaches the work to the holder of the “right”—the creator of the expression. Without a creator, who holds the right?
There is always a distance between the work and the creator. At some point, that distance becomes too great to confer the privileges of ownership. Pollack’s paint-can and brush shaker methods were deemed close enough—he created the machine with a specific outcome in mind (we can argue about how to describe that outcome). But the creators of the AI algorithms don’t have a specific outcome in mind—if they did, it wouldn’t be AI. The distance becomes too great.
But the case was, I’ll bet, not about the algorithm creator, but rather about the user who made the request. The case perhaps suggests that instructing a computer to make something does not confer copyright, unless the instruction is detailed enough (close enough) to provide the expression. The distance between the instruction and the product can’t be too great.
So, copyright is the same thing as the question of whether a photograph represents factual truth—the chain of creation has to be verified and verifiable to confer the origin of the image.
Rick “for whom photography is about the expression more than the product” Denney
Bookmarks