I wonder if such overriding divisions are just artifacts of our brains limited computational power, and whether it will survive when such limitations disappear.

Think for a moment about why and where we use `we'. I think that we use it as a heuristic to represent a group who are similar to us in some sense, ignoring the various differences individual members have, and I suspect we evolved such a notion for an evolutionary advantage.

Distinguishing members or environment which allowed for the best chance of survival or propagation of one's progeny gives one a certain advantage, and in the absence of computational power to precisely compute how best to do it, the simple heuristic of distinguishing a group based on some similarities could serve as a substitute. This heuristic holds some value even now, because we still lack the power to compute the relationships and advantages precisely.

But will it hold the same value in the future when the required computational power becomes available to individuals? And if its value diminishes, why should some entity feel the need to deploy this heuristic?