Ariel VA Feinerman

The Sibyl System, Psycho-Pass TV series

What is needed is nothing less than a breakthrough in philosophy, a new epistemological theory that explains how brains create explanations.

~ David Deutsch, British physicist and philosopher

Many scientists and engineers have unreasonably high expectations of artificial intelligence. For some reason, they believe that once thinking machines are created, they will immediately solve many of our problems and lead to a technological Singularity. However, this way of thinking is a misconception.

Note: I deliberately avoid using the word ‘intelligence’ because it usually refers to the ability to analyse problems and make decisions based on patterns, without awareness of the process. In this sense, intelligence is inherent in the Maple programme, which solves complex equations, as well as in bees, which are capable of learning and complex behaviour, and in many other animals. Instead, I will use the word ‘mind’ when referring to human-level thinking process.

Do we even need an artificial mind? Yes, we do. Unfortunately, ‘bare’ intelligence solves only a narrow class of problems, and sometimes quite inefficiently. Furthermore, ‘bare’ intelligence is incapable of formulating the problem itself! Take chess, for example: a machine solves the problem non-creatively, expending considerable resources and using various algorithms programmed into it by engineers, not to mention the enormous size of the training set. A human solves the problem creatively, generating new solutions while using far fewer resources!

1

Now, let us return to our problem. It is most likely that the first artificial mind will be be modelled on the human mind. This is because we still do not know how consciousness works, and the only way we currently have of constructing an artificial mind is to copy the human mind. In other words, the thinking machines we create will emulate the human brain. This is a very important circumstance that places serious limitations on our expectations.

The problem is that we are planning to set creative tasks for an artificial mind. We expect it to make breakthroughs in medicine and biology, physics and cosmology, sociology and psychology. Moreover, we believe that it will be capable of writing books, composing music, and painting pictures. In other words, we believe that it will be capable of creating new knowledge (in a broad sense, including art and music). But most healthy, normal adults do not engage in anything like this! They are indeed normal, because the majority determines the norm.

Being created on the basis of the average human, the artificial mind will not differ from it in any way, except for its potentially unlimited resources. It is quite possible that an artificial mind, possessing consciousness, simply will not want to engage in all this creativity that has been imposed on it in advance. After all, modern humans have unlimited resources compared to their Stone Age ancestors. They can access all the information accumulated by humanity in a matter of seconds. However, few people need this information, except for scientists and engineers. Most of the information technology works for entertainment. This has always been the case and will always be so, regardless of the amount of resources available (provided that human nature remains the same).

Therefore, creativity is more of an anomaly than the norm. Of course, consuming information is in itself a creative process that requires a certain amount of creativity, as does life in the modern world. Nevertheless, such everyday creativity is necessary for survival and differs greatly from studying the Universe or writing music.

For example, the advent of computer programmes for making music has made life easier for novice musicians, but it has not increased their numbers. If a person is not interested in music, he or she will not pursue it, no matter how easy the process may be. Similarly, no matter what computing resources a thinking machine may have, it will not solve a problem if it is not interested in it.

However, several ideas have been put forward on how to overcome such challenges. For example, an artificial mind could be programmed to solve specific problems. Nevertheless, until we understand how the human mind works — indeed, how any mind works — and what changes need to be made to the brain at a low level in order to stimulate an interest in music, we will not be able to influence our brains, and consequently their computer models.

Well, opponents would say that, while we cannot yet program the mind directly, but we can nurture the desired personality. This is certainly a more realistic idea, but it is still not feasible within the framework of today’s knowledge. (Moreover, nurturing itself is a form of high-level programming!)

2

There is another equally important issue that is rarely discussed, which is unfortunate. The problem lies in human nature. Humans are animals, and it is normal for animals to seek physical pleasure and reproduce. Anyone who spends their life creating new knowledge instead of earning a lot of money with minimal effort and seeking more physical pleasure could be considered ‘insane.’ In psychiatry, it is not considered normal for one idea to occupy a person’s mind for many years. Many talented people are indeed insane or have other mental health issues. Of course, such people enjoy their creative process, but the problem is that it can be dangerous to their mental health or even their lives. This is just as true of creative people themselves, who can sometimes be dangerous to those around them.

Our entire civilisation was built by ‘crazy’ people! Everything around us was created by scientists and engineers, artists and musicians, writers and other creative madmen who devoted themselves to changing the world, sometimes at the expense of their own lives! Yet we have more or less become accustomed to ‘mad geniuses,’ and it is easier to control a vulnerable person than a machine. Would you entrust your life to a powerful artificial mind if you had even the slightest doubt about its sanity?

In the near future, it is quite possible that we will learn how to construct truly creative thinking machines, but will they be as crazy as their human prototypes? And aren’t mental health issues an integral part of any creative process? These are questions to which we do not yet know the answers.

3

In other words, an artificial mind will face the same problems as a human mind. How can we avoid them? We need a universal theory of the mind that explains the principles of how any mind — or at least the human mind — works. Only by understanding these principles will we be able to design intelligent machines — not through blind copying or trial and error, but through purposeful, rational design. We will use the human mind as a basis, and either improve upon it or redesign it completely. With the help of such a theory, we can improve our own minds as well.

Compare the first clumsy aeroplanes with today’s civilian airliners or military fighter jets, such as the F-35! Modern aerospace engineering is a perfect example of our good understanding of the laws of physics. Engineers do not copy a bird’s wing; they calculate the necessary parameters using the laws of physics.

Alas, the development of artificial mind is slightly different: we have not yet a universal theory, nor have we attempted to construct one. This means there is still a lot of trial and error ahead, and a long way to go.

Bibliography:

David Deutsch (2012, October 3). Creative blocks. Aeon.

Ariel VA Feinerman is a researcher, author, and photographer, who believes that people should not die from diseases and ageing, and whose main goal is to improve human health and achieve immortality. If you like Ariel’s work, any help would be appreciated via PayPal: arielfeinerman@gmail.com.

Read more about Ariel VA Feinerman on the THPedia entry about him here.

Ariel VA Feinerman was the guest of honor at the U.S. Transhumanist Party Virtual Enlightenment Salon of  May 10 / December 7, 2025.