Stephen Hawking wants to ban AI military robots. How close are we to having actual "killbots?"
In my story "A Star To Steer By" the autonomous weapon is a spaceship. In Tom Kratman's novella "Big Boys Don't Cry" he paints a grim picture of the fate of sentient tanks. (Both stories are very similar with exact opposite endings and different points. Mine is rather more optimistic).
How about real military robots?
How close are we to this kind of technology?
Right now, military robots tend to fall into two categories:
The first is drones, which are remote controlled by a pilot. They're cheaper than manned planes and pilots are, well, very expensive to train. So, despite the slight time lag, remote controlled planes are taking over from manned fighters.
The second is hazardous environment robots - used for bomb disposal, mine seeking and search and rescue. These are also remotely controlled by a soldier...and demonstrate an interesting tendency. Soldiers tend to name these robots, get attached to them and get upset when they get blown up. They're treated much the same way as the dogs (and rats) they sometimes replace.
The navy is working on robots to fight fires on ships - again so that less expendable sailors don't have to.
The theme is obvious: Military robots are used to protect human personnel by going into dangerous situations so they don't have to.
None of these robots can yet operate without a human controller - although there's some progress with drones that can identify targets for assassination.
What's the big worry? The worry is that the more robots fight our wars for us - the more likely we are to go to war. And, well...it's not like nuclear weapons.
Take Iran, for example. Iran just unveiled a robot tank, the Nazir, which appears to be armed with either a machine gun or a pair of man-portable anti-air missiles. Again, operated by a remote driver, but...oh, yes, and it can be smaller than a manned tank, too.
Syrian rebels have already used remote controlled gun robots.
And, you know, why sacrifice a person in a suicide attack if a robot can do just as well?
In other words - Hawking and his friends have a point. On the other hand, I think the robots in war ship has thoroughly sailed.
Which means that the issues raised in A Star To Steer By and Tom Kratman's Big Boys Don't Cry are going to be real issues we may have to face soon: When our robot soldiers start becoming sentient, what do we do?
Will robots make war more likely? More to the point, will we treat sentient soldiers as tools to be conditioned, used and then rejected (Big Boys Don't Cry). Or will we treat them as people...but still people with a purpose, who don't have rights outside the military (A Star To Steer By).
Will they be tools, slaves, or will be have the strength to realize that there is a point at which they have to be treated as soldiers. And allowed a soldier's freedom to say "Enough" and go home.