We can only be confident of that if they're in this universe. Also, this is only a conjecture. It's just something some of us imagine would be true. You can't elevate such an idea to the level of empirical data.
So, does the fact that something is discovered (rather than invented), and can be discovered many times independently in identical fashion, confer it some sort of elevated ontological status?
Unicorns. Kirin. Narwhals. Rhinos. So you have two independent origins in culture for such a creature, plus two appearances of something similar in nature. Does that confer it some sort of elevated ontological status?
It's not at all clear that the cultural constructs are independent of each other or real world creatures. Particularly in the Western case, some of the earliest Greek accounts of the unicorn as well as Marco Polo’s account confirming its existence and difference from the contemporary evolution of the image stem from the distant East (from the European perspective) and clearly are describing a rhinoceros. They are also very similar to accounts of the Kirin.
Rather: if we have a notion of something and statements about it in every universe, then it must exist. However, come to think of it, that's bonkers.
Still, it's remarkable that mathematics allows us to make reproducible true statements about fictional objects.
What would happen to your epistemology if every alien civilization has unicorns but some lack π?
I knew it, Unicorns are real!!11!
I am not sure I can conceive of aliens that don't have pi (how can they have cylindrical rockets?).
Anyway, maybe we should let the professionals deal with this.
https://plato.stanford.edu/entries/platonism-mathematics/
https://plato.stanford.edu/entries/intuitionism/
https://plato.stanford.edu/entries/mathphil-indis/
Good question, but I think there might be a difference. Suppose, in a galaxy far, far away, there are aliens. They probably don't have the concept of unicorns. However, they probably have discovered numbers, pi, Pythagoras theorem, e^(i pi) + 1 = 0, etc.
So, does the fact that something is discovered (rather than invented), and can be discovered many times independently in identical fashion, confer it some sort of elevated ontological status?