Prerequisites for the emergence of AI superior to the human brain:

  • Moore's law for quantum computers;
  • the emergence of programming languages ​​for quantum computers;
  • a quantum computer is similar to the work of human intuition, will and consciousness - almost instantly sorting through all possible solutions to the problem and choosing the optimal answer;
  • the prohibition of cloning quantum states is similar to the impossibility of cloning our consciousness;
  • quantum AI should complement classical AI, just as in the human brain its various structures work as a single mechanism;
  • any classical AI is trained to solve only one problem and works more efficiently than a person only in a narrow segment of activity. For example, a trained AI model plays chess better, but is not capable of doing anything else. Quantum AI should replicate approaches to solving various types of tasks, speeding up the process of learning a new task.

Suppose.


The permanent "I" of man does not exist. The “I” of an individual arises in response to a person’s interaction with the world and is necessary for him to work more effectively with the world around him. In different states, a different “I” arises: in a dream - one “I”, in reality - another “I”.

Therefore, there is no fundamentally insoluble problem of transferring consciousness to another medium.
Unconsciously, the most automatic brain elements that control the functioning of vital organs work in a person. As soon as the need arises to make a decision and make a choice, they talk about the appearance of a certain "I" - consciousness. The choice is made by the element of the human brain, apparently having a quantum nature - an analogue of quantum AI. True, the presence of a conscious choice does not necessarily mean the emergence of self-awareness - there is no one "I". For the appearance of “I”, active interaction with memory is required - with that memory that can be visualized, heard or virtualized (represented) in some other way, accumulated by human experience. Moreover, various blocks of the content of the experience, access to which is carried out, for example, in a dream or in reality, gives rise to different "I" of a person. Although sometimes part of the experience of various states penetrates each other, and thus one “I” of an individual combines with another “I”, enriching each other.

image

Transfer of human consciousness to a machine carrier


The transfer of consciousness, apparently, will be carried out as a trend appears on the integration of neural interfaces into the human brain. At first, this will be relevant for increasing human capabilities, and then for transferring consciousness to another medium, either autonomous or into a single network - the quantum Internet. As the power of the implanted element grows and a quantum AI appears in its composition, a second consciousness and “I” will appear in the human mind, which will be activated at certain points, for example, in critical situations, when the human mind will not be able to quickly solve the problem. Then there should be a merging of the two types of consciousness and the emergence of a single "I" with an mutually enriched memory that will be able to choose one or another experience. In the future, it may be possible to disconnect one type of consciousness (for example, when the human body dies) and complete self-identification with another type of carrier without loss of accumulated experience and continuity of self-awareness (or its restoration after "awakening").

The rapid development of AI and the advent of a quantum computer, the technology of which is already developing according to Moore's law, raises the question of the emergence of consciousness in a computer. For a long time it was believed that computer self-consciousness (in the sense of awareness of one’s own actions and the ability to make decisions independently) can arise after the AI ​​has developed and become more complicated. However, not all brain functions can be translated into classic AI. A neural network can learn one skill and it seems that it is not able to transfer its experience to other areas of activity. In addition, in problems associated with a large number of solution options (with enumeration), the time to find the optimal solution is extremely long, which is not comparable in efficiency to our brain. T.O.a classic computer is probably only part of a complex “device”, which necessarily includes a part working on quantum principles. As the human brain develops, a quantum leap in its evolution, separating humans from other animals, is probably associated with the appearance of that part of the cerebral cortex, whose work is based on quantum principles.

Modern AI technologies have followed roughly the same path as the evolution of living things on Earth: chemical networks first appeared, and then faster - neural networks, but still performing a single function. The vital part of the information was recorded in a genetic sequence so as not to repeat the training in the life process of one organism each time, but to pass on “best practices” from generation to generation. At some point in evolution, a quantum leap occurred, when quantum parallelism appeared in the human brain, which allowed us not to learn each skill sequentially, but to analyze, generalize and find the best solution from the point of evolution almost instantly.

We call consciousness a certain “higher” brain function, which makes a choice, makes a decision. Consciousness is directly related to mental activity, forecasting and modeling, assessment and analysis of the situation, imagination and choice, i.e. this is a complex multidimensional brain work and it is associated with enumeration and analysis of all possible options. Consider a simple example - a person wanted to eat a piece of cake... At some point in time, the person began to lack sugar and he sends a signal to the brain - the thought arises: "I want a cake." In fact, the mind interpreted the original need: "Need sugar." Why does this process go through the stage of awareness of this need? It would be possible, bypassing the conscious part of the brain, to send a person to the refrigerator and get a cake from there.... As it were, a person is given the right to make a final decision: whether or not to eat a cake. At the same time, it becomes possible to simulate the situation, predict the possible consequences, determine its importance, rank the needs, determine how timely it is now, etc. - i.e. think about it and make a decision.

For the vital processes of the body of the body (heartbeat, respiration, digestion, etc.), consciousness is usually not integrated into the decision-making process, i.e. information is not transmitted to the conscious part of the brain. These processes are already defined (automated) in the process of evolution and conscious intervention is no longer required. Therefore, we can assume that consciousness is needed there (and it is built into the information chain to make a decision) where evolution has not yet worked out the optimal way of functioning.

In our brains, two components can be distinguished that are similar to the two types of existing computers: classical and quantum.

In order to understand which parts of the human brain are based on quantum phenomena, we first denote the classical ones. Obviously, the first ones on this list will be the parts of the brain responsible for processing the signals coming from the sensory organs: an analogue of AI for pattern recognition. Signals from the outside world are sent to the input, at the output a “picture” is an image of an object with which other parts of the brain work. Some signals are processed by the unconscious - automatic elements of our computer. The entire ancient part of the brain - areas associated with the vital processes of the body - is a classic computer, an area of ​​the unconscious, developed by millennia of evolution.
Others go through the latest evolutionary brain regions associated with the work of our consciousness, which appear to be analogous to quantum computers. Consciousness is embedded in the information chain for decision-making only in the process of initial training in new skills, i.e. the first few hundred times, depending on the types of skills. So we learn to drive a car consciously. But after mastering the skill “before automatism” we completely transfer this function to the unconscious. After training, conscious decision-making and the control function is fully automated and goes to the unconscious level - that is, a new skill is being formed. At the same time, the speed of decision making increases many times - the more skill development, the higher the reaction rate.Examples of training may include: learning a language or painting, driving a bicycle or car, teaching dance or martial arts. When new information appears with a skill that has already formed, the conscious function is activated again, training our neural network, and then it goes out again. It is characteristic that the perception of time during learning or other conscious activities is lengthened (for example, traveling in unfamiliar places), but on the contrary, subjective time is shortened for routine operations.
(E. Schrödinger wrote about the emergence of consciousness in the learning process in his works)

The classical neural network of our brain learns only one skill and does not affect the operation of other neural networks. A quantum computer does a good job of sorting out values. The quantum network of our brain allows you to instantly sort through all possible options and thus is able to optimize the thinking process, choosing the best solution to the problem. Our brain first creates a virtual set of images - a model of the future - and carries out quantum sorting through them. The search itself is carried out due to the phenomenon of quantum parallelism and occurs almost instantly.
When training, attention is focused on one task and the better the concentration of attention, the better the result of this training. As a rule, people with good attention achieve better results. The level of motivation also affects the speed of learning through the mechanism of attention (concentration).
The process of training a neural network in itself is the selection of the optimal parameters of individual neurons to obtain a better response to incoming signals.

In what way, and most importantly, why does consciousness enter into the learning process - a quantum computer of the brain? Obviously, better coping with the tasks of enumeration, he better models the future situation and, without waiting for feedback from the outside world, goes through all possible situations that could arise in the real world. Training time is significantly accelerated, increasing the survival of the body. T.O. natural selection chooses conscious activity.

There are still many questions about the quantum nature of human consciousness, in particular, what material carrier provides quantum consciousness in the human brain? This is a topic for future research.

The best proof of the hypothesis of the quantum nature of consciousness or its refutation will be the creation of AI based on a quantum computer.

see Neurons of consciousness E.N. SOKOLOV .

Source