Autonomous Attack Aircraft

For my disruptive innovation case I had the topic of autonomous vehicles. In an article I found on justsecurity.org, Michael Horowitz and Paul Scharre discuss three areas of concern related to fully autonomous aircraft. First area of concern is related to accountability. If autonomous aircraft fail to do their job properly, who’s at fault? Another issue is the off-loading of moral responsibility for killing, which could lead to unnecessary deaths. Finally, if an autonomous aircraft became out of control, there could be serious damage inflicted to innocent people or casualties to friendly fire. Because of these concerns, many people advocate a level of human control over these aircraft because humans can make better informed decisions for a particular situation. Because of this, there is a debate over what is the right amount of autonomy and human control.

How much human control do you think is necessary and what functions should be autonomous in an attack aircraft? Or, do you think it is possible to have a fully autonomous attack aircraft? If yes, how would you mitigate the concerns mentioned above?