about the project

---

The concept of artificial intelligence first appeared in 1956 after the so-called "Dartmouth Seminar" - a conference on artificial intelligence at Dartmouth College. Though the ideas of a machine or artificial intelligence existed long before that. During the Second World War, British mathematician Alan Turing created a machine for reading encrypted messages, which, with a certain stretch, can be called the prototype of artificial intelligence.
AI-1 - precise algorithm and technical neural network
The initial idea was to replace a person by creating a digital computer mechanism for this with a finite and initially predefined set of working algorithms. On the basis of this approach, the idea of creating a machine capable of simulating a plurality of neural connections with the help of an artificially created technical neural network appeared. This was the first generation AI.
AI-2 - focus on information processing instead of creating a rigid computing mechanism
In the late 80s and early 90s, a new concept of AI emerged that could be called the second generation. At this stage, the work of the AI-2 was supposed to be based on the processing of signals and information using flexible algorithms. Since then and to this day, most of all companies and research groups have been working in this direction, which can be called AI-2. By the way, search engines work exactly according to this principle.
AI-3 - machine learning
Finally, at the beginning of the 21st century, along with the growth of computing power, a new concept of so-called machine learning emerged. The essence of this new model of artificial intelligence, let's call it AI-3, boiled down to analyzing data associated with a phenomenon and a process and determining the dominant trends in making successful decisions in the past, in order to create current active control algorithms in the future. In addition, the predictive capability was assumed based on this training.
AI-3 + or Deep Learning and AI-3 ++ Enhanced Deep Learning
It should be noted that the embryo of AI-3 and AI-3 + existed even at the stage of AI-1, but then the low power of computing systems did not allow achieving real results. In the second decade of the 21st century, thanks to the use of computing systems of increased performance, it became possible to force the machine to learn using huge databases and multilayer technical neural networks. This process is called deep learning and can be called an advanced version of AI-3 or AI-3 +. To see the difference between AI-3 and AI-3 +, you need to look at how the neural machine translation system created several years ago works. When creating a new system, the developers completely abandoned natural language analysis algorithms and replaced them with a simpler model using exclusively comparative analysis of finished translations. The new simple method impressed with its accuracy and efficiency. Instead of unraveling and clarifying logical connections in the structure of natural language, the new approach proposed a model in which all processes proceeded according to the principle of a black box. A similar technique was used by another company to create an Artificial Intelligence capable of fighting any professional go player. This AI uses two neural networks at once, working on the principles of AI-3 + (deep learning), one of which analyzes the position, and the second the strategy used in the protocols of past parties. This approach, which we will call AI-3 ++ or enhanced deep learning, is the most advanced concept in our time. Various companies are continuing to create new systems based on this principle.
All the systems described above are unitary complexes that, using electronic computing algorithms, try to form the most successful response to the received request. Their weakness is that real intelligence does not work on a unitary principle, but is a double system that functions as a classical symbiosis in biology. For this reason, "Cynt" (AI-4) has no competitors, since there are no systems in which the brain and the machine work together and within a double system.
The territory of new fundamental science
This is an uncharted technological and scientific area and we will be talking about a completely new fourth-generation AI system. To date, there are only a few attempts to implement such projects. The weakness of these attempts is that they do not take into account the dual system of organization of human intelligence, and therefore do not understand the difference between the reflex and intellectual mechanisms of initiation of nerve impulses in the neurons of the human brain. What they think is a single system is in reality a complex organized symbiosis to connect to which is impossible if one does not understand how it's individual (constituent) components interact with each other. On the other hand, understanding the connection scheme, it becomes clear that it is possible to create an integrated system without invasive and traumatic interfacing. 

Nowadays, many people are engaged in artificial intelligence. This is almost always done by technicians with varying levels of knowledge about how real intelligence works. Our position is that in order to create something artificial, but at the same time truly working correctly, you must first understand how it works in nature. It is for this reason that we have created our own physiological concept of human intelligence, based on a whole system of well-known and scientifically confirmed facts. This is a work that took almost 30 years.
Our first advantage is our own concept in neurophysiology and quantum physics
Now we have our own theory of the dual-functional system of the human brain, which explains how signals of two types are formed in the neurons of our brain. We have worked out our concept to the level of understanding the mechanism of initiation and transmission of a nerve impulse in each specific synapse and can describe it from the moment each ion channel opens on the postsynaptic membrane of a neuron in the synapse zone. This model required us to understand the mechanism of quantum interaction of particles that make up the structure of ion channels, which has become a separate concept in the field of quantum physics. All this as a whole gives us a clear and understandable basis in the form of describing the process of the human intellect from the point of view of fundamental science.
The second advantage is the interface of its own design
A common problem for all projects trying to connect the human brain and the computer is the complexity of arranging the signal transition point. Physically, this is the place where the signal from neurons goes to the inanimate carrier and vice versa. The main problem here is not even that any clearly and accurately working system will be very traumatic for biological tissues, the main difficulty is the interpretation of the received impulse. We were able to create a model capable of circumventing both limitations. Our system is extremely minimally invasive (not traumatic) and at the same time very accurate and in detail interprets the signal. For this, our interface will be made of special material that is not known and is not currently used by any of the competitors.
The third advantage is a special functional diagram of the complex
A deep understanding of the presence in the structure of the human brain, in fact, two independent, but at the same time very closely related systems gives us the opportunity to attach a third artificial component in exactly the place and in such a position where it is most effective from the point of view of physiology. We not only have our own connection interface, but we also know where and how it should be located in order to work with a living brain in pairs. As a result, this greatly facilitates the task for the software package, which, due to the correct configuration of the elements that make up the system, will save time and resources when making decisions. This will allow our system to be several times superior to any fully electronic artificial intelligence system in speed and performance.