We've never seen randon mutation and selective pressure ever really build anything, only trigger and exploit existing adaptive rulesets... pushed to extremes mutations may often yield a benefit in a narrow situation but at some greater cost, like throwing the backseats out of your car helps you accelerate faster, but reduces the overall utility and flexibility of the use of the vehicle. Im sure if we started honestly looking at organic systems as the product of thought, it would yield a greater understanding of them that we would be able to leverage in industry.