The cost of a mistake may be lower in the case of cuteness than sexiness.

"May" be. I don't see this as being demonstrated. I prefer explanations like Eliezer's that show greater selection pressure - there's a whole range of explicit sexual selection, but no real selection on other species for cuteness.

Here's another explanation: other species don't benefit from being cute-to-humans, so they don't spend their time trying to cheat humans into perceiving them as cute. But humans are deliberately trying to be sexually attractive and are very good at taking advantage of any weak points in our sexual heuristics. Therefore our heuristics evolved to eliminate false positives.