Perhaps the easiest answer will come from the way that we treat them
BY THE middle of this century, machines could be demanding the same rights as humans. If granted, nations would have to provide benefits including energy, housing and even 鈥渞obo-healthcare鈥�.
When these remarkable predictions were made in , robotics experts were not impressed. There are more pressing concerns, they said, such the safety and legal liabilities of robots, and the increasing robotisation of the military.
Advertisement
Machines have got much smarter in the four years since that report was published. In 鈥淏ot shows signs of consciousness鈥� comes the latest news of a computer program that was inspired by global workspace theory, a popular model of human consciousness.
When Stan Franklin of the University of Memphis in Tennessee unveiled the first incarnation of the machine, to assign jobs for the US navy, a colleague suggested he claim that the program had subjective experience. Franklin declined. But he has since created LIDA, an upgraded version that can learn. This program matches people in the way it reacts and perceives when carrying out two simple tasks, providing a tantalising hint that it may indeed be a useful model of human cognition.
Right now the software bot doesn鈥檛 interact with humans, or occupy an environment sophisticated enough to warrant a sense of self. Even when it does, figuring out whether LIDA is truly conscious is what philosophers call a 鈥渉ard problem鈥�. As they delight in pointing out, there鈥檚 no way to tell that you aren鈥檛 the only conscious being in a world of zombies.
A more basic issue is that there is no agreed definition of consciousness. Perhaps in practical terms, a simpler answer to the question of machine rights might come from the way people treat them. We should put our faith in our own ability to detect consciousness, rather than look to philosophical discourse.
聯We should look to the way people treat machines and have faith in our ability to detect consciousness聰
There is one obvious shortcoming of this approach: we will probably sense sentience before it is truly deserved because of our remarkable tendency to anthropomorphise. After all, we are already smitten by today鈥檚 relatively dumb robots. Some dress up their robot vacuum cleaners. Others take robots fishing or go so far as to mourn their loss on the battlefield.
Even so, popular sentiment towards machines and robots will give a vivid feel for the degree of their sophistication. Franklin himself admits that even he referred to his original creation as 鈥渟he鈥�, though he 鈥渄id not feel at all bad鈥� when he turned 鈥渉er鈥� off. But when he and a significant number of others do feel a pang of guilt as they flick the off switch, we might well have passed a milestone in artificial cognition: the birth of a machine that deserves rights.