Machines and technology have been man's best allies for many years now. Although in the beginning machines were used to make easier for us the most challenging and rough jobs, today they have spread to almost all areas of our lives. Moreover, as we develop as a society, machines accompany us in this process, creating ever more sophisticated artefacts.
Recently, we have been hearing about the Artificial Intelligence (AI) of machines and the impact it could have on our future. Artificial Intelligence is the combination of algorithms designed to create machines that have the same capabilities as humans. Although it may sound like science fiction, it has been a permanent part of our daily lives for some years now. For example, AI is present in facial detection in mobile phones, or in voice detectors such as Siri or Alexa.
With AI, machines are moving from being passive tools that we have to manipulate to get results; to being generative tools, capable of carrying out activities on their own once we have given them the objective and certain criteria.
If we combine AI with creativity, the results could lead to new designs that we would never have thought of. The process would be as follows: we give machines certain characteristics that we want our design to fulfill, they explore all the possibilities that can result from our criteria, and finally, they give us a design that we would probably never have been able to imagine, and that fulfills our previous wishes in the most efficient way.
If we combine human activity, Artificial Intelligence, and the activity of robots capable of precisely executing what the human explains, we would be able to carry out new impossible projects. Robots would help us to build and AI would help us to imagine.
On the other hand, there is a certain wariness about the unstoppable advance of the technology, and it is important to be aware of certain precautions and limits that must be set around it.
For AI to be compatible with human life in the future, according to computer scientist Stuart Russell, robots must comply with some principles:
Their sole purpose is to perform their activity following human values.
The machine will have an uncertain idea of what these values are.
Human behavior will provide the necessary information to the machine about these values.
In other words, the AI machine will have a goal, but it will never be concrete.
This idea may seem a bit confusing, and Stuart Russell explains it with the following example: if we provide an AI robot with the concrete goal of making coffee, its logic will lead it to think that it must fulfill this function above all else. Thus, it may conclude, "if you turn me off, I won't be able to fulfill my goal", so that the robot itself could disable the switch-off function, and we would lose control over it. Russell, therefore, proposes that the robot should always perform its function according to the ethical values of the human being, which, not knowing for sure what they are, the robot may think that at a certain moment it could be necessary for someone to turn it off. Thus, the uncertainty of the robot ensures that we have control over it.
Robots would learn to reach their objectives more efficiently with experience, i.e., when someone shuts them down, they may learn that what they were doing may not have been the right thing to do.
Finally, the robots will not be designed to satisfy the user's preferences, but those of everyone, trying to create altruistic machines.
As conclusion, I would like to add my personal point of view. Despite the facilities that AI can offer, which I find absolutely incredible and advantageous when it comes to finding solutions to certain problems that affect us all, such as climate change, I am not so sure about the benefits of its intrusion in some tasks of everyday life. We live in an increasingly individualized world, partly because of technology. We interact less and less with others, or find it harder to do so if not through a screen. So it scares me a little to think of a future in which robots intrude into our daily lives, replacing us in the execution of tasks that may involve the creation of close bonds between people, such as caring for children or the elderly. I have never been a person of strong convictions, and my current rejection is not free of doubt, so perhaps in the future I will look at the development of AI from a different point of view and think the opposite of what I think now, but for the moment, its fast advance scares me, as does the idea that this future is not so far away.