via statistical imitation. other methods, such as solving and implementing by first principles analytically, has not been shown to be np hard. the difference is important but the end result is still no agigpt in the foreseeable and unforeseeable future.
there is not a single thing that could wipe out a deep sea habitat that wouldnt also wipe any space colonies. but i dont see anybody arguing for that, despite being far more achievable and practical. also, there is no feasible way for space colonies to be self sufficient anywhere in the near future, so wiping out earth also wipes out space colonies relying on it for supplies. this argument aboOt survivability is absurd.