I can think of a couple of counterarguments to both points (and that part of the text in general): 1) one thing that is often overlooked is how hard is to pinp

I can think of a couple of counterarguments to both points (and that part of the text in general): 1) one thing that is often overlooked is how hard is to pinpoint the carrier and provide this information to a missile. For a sea-skimming missile horizon is at ~12 miles (20 km); given the distances involved (F-35C has a combat range of about 1k km) and carrier's speed (60 km/h) you will need to provide updates to a missile regarding current carrier's position, which you can't know if you aren't over the horizon for a carrier — in which case you will probably be shot down. This makes the danger of low-cost missile platforms somewhat lower. For a more thorough discussion of the problem see [1]. 2) Drones are unusable against sophisticated opponent. First of all, there will be no datalink: space assets will be denied (either jammed or destroyed), direct radio links will be jammed, too. Therefore the drone will need to orient, fight and decide who to kill autonomously, which is problematic both from technical (e.g. autonomous navigation and decision making) and moral (autonomous killing machine, anyone?) standpoints. Therefore, we can't switch human pilots for drones. [1]: http://lexingtoninstitute.org/wp-content/uploads/aircraft-carrier-invulnerability.pdf http://lexingtoninstitute.org/wp-content/uploads/aircraft-ca...