That astronaut with a coffee cup in space still looks very silly to me, apart from the eyes, just because the cup handle is facing the wrong way while the hand is positioned as if holding it by the handle.
We're working on ways to evaluate if an image looks "off" - but its a hard problem because a lot of it is subjective
I have no clue about SD other than toying with it, so forgive if this is naive, but do you have any plans to maybe feed this back into the generation, e.g. jiggle the the seed or other parameters around a little until no fingers are malformed?
I do think over time the feedback loops will look much different, as the models get more solid.