The interesting thing about it, is if you were to imagine yourself performing an experiment on the evolution of a population to determine which traits lend toward survivability, to simulate what you're proposing, the population within the constraints you defined, would in effect attempt to subvert the experiment by changing the constraints to suit its collective preferences. It would ruin the experiment in so far as you wouldn't have found those traits that improve survivability within constraints that matter, and given that we in the real world have no such controls over the real constraints that matter for the survivability of the human race, our own attempts to guide human evolution are similarly self-deluded, selfish and shortsighted. That having understanding of evolution somehow allows us to control our own evolution, seems to lead to a kind of contradiction wherein we seem to think that subverting evolution is evolution. It's fundamentally flawed.

I suppose none of that ultimately matters when there are untapped markets to explore with designer babies and what not. Or maybe it does, when at the end of the day, everything we do is factored into selection of the fittest whether we like it or not.