Northrop Grumman's X-47B operates autonomously with no one in the cockpit and no one on the ground in control either. Tests are ongoing to determine if the drone can land on an aircraft carrier deck completely on its own guided by its own control mechanisms.
This is just the beginning of research that will probably end with drones who will themselves make the decision whether they should destroy targets they are programmed to seek without any human on a computer somewhere deciding what should be done. Humans will figure in the loop only as targets not as participating in the decision to fire missiles.
Jacob Kellenberger head of the ICRC (International Committee of the Red Cross) noted that the capacity to discriminate in attacks is required by international law. Autonomous drones' capacity to discriminate will "depend entirely on the quality and variety of sensors and programming"
There are no plans at present to allow autonomous drones make decisions whether to fire on chosen targets. An Air Force report states:"authorizing a machine to make lethal combat decisions is contingent upon political and military leaders resolving legal and ethical questions." If a legal justification is required then no doubt legal minds can be found to come up with the proper opinion. However, this will no doubt be some time after the military or CIA has already gone ahead and used the drones. For more see this article.