You know more of what a giraffe is than DALLE; you can imagine one as a 3D model, for instance. It can't do this, it can only render different 2D images of one it's seen in photos. Of course, this is still impressive that it knows how to remove the backgrounds from those.

Examples of prompts that don't really work are "spherical giraffe" and "the underside of a giraffe".

We'll need more 3Dness to generate other media like models and video, so it should be coming.

I agree. One thing I've noticed with most image generation, including faces, is that it makes the same mistakes that a novice artist makes. It draws eyes where eyes should be, mouths where mouths should be, but doesn't actually shape them correctly for the 3D head and viewer perspective. It can perfect a specific viewpoint with lots of training (like a novice artist who perfects drawing realistic faces and not much else). But my interpretation of much of the artifacts are that they are just missing structural cues.