Recently I read an article on TechCrunch titled: Technology can’t replace the human touch by Ron Miller, and it made me think. Can technology, or to be more specific, intelligent technology replace the human touch?
It is natural to think that humans are special. That there is something about us that cannot be replicated. It is a philosophical question, whether this is indeed the case. Some people even see this as a spiritual question. Can our brains be replicated in such a way as to preserve or create an individual real personality? What does that even mean? At least it is certain that those questions are not going to be answered by me.
Is the human touch special?
But there is this other question. Can intelligent machine replace what the TechCrunch article calls The Human Touch? I would like to explore this question a little bit. Also, there is one kind of people who have sometimes elaborate visions on what intelligent machines can do and how humanity will react on them: Science Fiction authors. Lets start by looking at the following quote from the article:
While a machine can perform a given task, often more efficiently than we can, what it lacks is the artistry in the activity, that uniquely human ability to cater to the needs of the individual.
What is artistry? What are the needs of the individual. Take the android Data from Star Trek, for example. At some point he performs a certain classical piece on violin. In order to do that, he combined several recordings from great violinists in the past to create a new performance, that had great influence on the audience. There was a certain need that he fulfilled here. Is this artistry? According to captain Picard it was, since he could have picked other recordings. Also in the whole Star Trek TNG series, the crew develops affection for Data and he is a fully accepted crew member in every respect.
Compare this to the fluffy robots for kids, or the attempts to create a conversational robot to help elderly people feeling less lonesome. They are a far cry from a real intelligent robot, but it seems to me that humans are actually able to derive some value from those machines on an emotional and personal level. And with respect to art, what if we are not able to detect who the author of certain artworks was? At this moment AI algorithms can paint in the style of a master, or develop dreams. What if they learn at some point to surprise us?
Ok, let’s look at another quote:
Even if a machine could determine an appropriate plan — and as we know there are few absolutes in medicine — we still want to work with a doctor, who has been trained to talk us through the options and administer the treatment protocol — and who understands that art in the science.
How much of this effect is real, how much of it is specific to our generation? When we look at self-driving cars for example. We don’t trust them yet. We want to drive ourselves. However, the next generation of kids, who might never have seen a non-self-driving car, would they have the same distrust and longing to drive themselves? Would these kids, when they grow up, care whether a machine performs surgery on them instead of a doctor? Especially if the treatment by a machine is more efficient, less painful (by being less intrusive), and more safe?
Can humanity do without intelligent machines?
Let’s look at one more quote from the conclusion of the article:
[…] but some things remain fundamental, and people-to-people communication will continue to be one of them.
In order to discuss this point well, I will tell you first about the science fiction work of Isaac Asimov. There are several books that span the future of mankind from the first settlements to tens of thousands of years later. Let´s summarize a few key points from this story to you.
In his story line, humans first settled on fifty new worlds. In order to assist the settlement, they were equipped with robots who possess a varying degree of intelligence, based on the function they need to perform. Robots keep a distinctive exterior and are generally regarded as a kind of lesser species. Because of the three laws of robotics, that make every robot protect human beings at all costs, this is never a problem.
At some point the robot Daneel is created. He has a real human appearance and is virtually indistinguishable from humans. Then people start to get feelings (of all kinds) towards him.
However, being pampered by robots, who take care of all needs (also medical needs) the fifty settler worlds become complacent. By eradicating all germs and excellent health care, they are able to extend their live span up to about three or four decades. This makes them risk averse and they do not explore the rest of the universe. In the end, there are new settlement waves from Earth which colonize the Galaxy further and based on the bad experience, robots are banned from society.
One of these fifty worlds, called Solaria, is particularly interesting. They have even more robots than the others. They don’t get extinct like the others, but they develop a taboo on person to person relations. So much so that they even employ bio engineering to make them hermaphrodites, so they don’t even need to meet to procreate. The Solarian robots are actually hostile to humans, since the definition of ‘human’ is changed for them to mean someone with a Solarian accent.
What is interesting in Asimov’s vision is that in the end the Galaxy is colonized without the help of intelligent machines as humanity has learned from these failures. However in an interesting twist, it turns out that the humanoid robot Daneel is actually the custodian of the whole Galaxy. He manages to steer the evolution of the Galactic Empire on his own via telepathic capabilities. It seems as though while Asimov foresaw problems with running our lives with intelligent machines, he also did not trust humanity to save itself. When one looks at how well we do on environmental issues, global peace, and so on, we might well doubt our ability to survive.
We will antropomorphize them
What is common to these science fiction stories and others like ‘Her’, ‘A Space Odyssee’, etc. is that people start to form a relation with something that is not human. We have the tendency to antropomorphize many things and we try to connect to things even though we are aware they are not living beings.
And intelligent machines become more and more realistic. Speech synthesis is becoming so real that it becomes indistinguishable from real persons. And what to think about an article called Robot gains Social Intelligence through Multimodal Deep Reinforcement Learning? The article describes how a robot learns by trial and error basic social interactions. How different are these things from us?
Is this good or bad?
When intelligent machines are indeed able to take over most of the interactions with us, is this a good thing or a bad thing? How can we retain our human nature? Should we? And when the machines perform most of the work, what are we going to do? These are all questions I do not have the answer to, but it is important to address them in the near future.
Technology is probably able to replace the human touch entirely. It is our task to decide whether it should. And if so, when we should or should not intervene? In my opinion it is best to create options at this point. We continuously advance in automating our society, this is unavoidable. But we should retain the possibility to override. There are after all many science fiction examples of intelligent machines that go bad.