Let me ask this then. What's the difference between what's being done with AI image generators (I won't call what they make art), vs a human artist using an existing image as a reference? The latter never ask either. If it's ok for the latter to do it with a few images, then the overwhelming issue seems to be the scale of the number of images used to train the generators. I don't think that's enough to make one ok and the other not ok.
Let me ask this then. What's the difference between what's being done with AI image generators (I won't call what they make art), vs a human artist using an existing image as a reference? The latter never ask either. If it's ok for the latter to do it with a few images, then the overwhelming issue seems to be the scale of the number of images used to train the generators. I don't think that's enough to make one ok and the other not ok.