First stage in the integrated Decisional System

The integrated Decisional System, alike any other previous (specific, standardized, particular) Decisional System, or even alike any other intelligence, system, or program, is formed by the traditional three stages that I am developing for all intelligence, system, or program, the three stages of: application (database or matrix as first stage), replication (the replication of all those human skills necessary to carry out its purpose), auto-replication (auto-improvement or auto-enhancement by itself).

In this case, the integrated Decisional System, the first stage is the integrated database of decisions, where are filed all global decisions in addition to all particular decision sent by the particular programs, and is the place where to carry out the first assessment: quick rational check for quick decisions, or first rational adjustment. The second stage is where the integrated Decisional System is going to project all decision: starting with the single project, going on with the global project (the plan), the actual project (the actual plan), and the prediction and evolution, virtual or actual, plan; and across all these projects the rest of six rational adjustments, plus the seven rational comparative adjustments. And finally, the third stage is where all those decisions on the mathematical projects, having passed all the assessments (quick rational check for quick decisions, seven rational adjustments and seven rational comparative adjustments), are transformed into a range of instructions.

In this post what I will develop, among all these three stages, is the first stage of database of decisions for the integrated Decisional System, identifying what decisions stores, how is going to manage the database, including the elaboration of lists of decisions, and logical sets of decisions in order to work easier with diagrams of Venn.

But firstly, a very brief summary about what the integrated Decisional System is, and what place it occupies within the chronology for the construction of the Global Artificial Intelligence.

As I have explained in previous post, the construction of the integrated GlobalArtificial Intelligence is a long process starting with the first Specific Artificial Intelligences for Artificial Research, by Deduction and by Application, followed by a second phase based on the collaboration between them, as experiments about how to build for first time in the third phase the standardized Global Artificial Intelligence, as a synthesis of all specific matrixes coming from all the Specific Artificial Intelligences for Artificial Research by Deduction, followed in the next fourth phase by the Unified Application as synthesis of all databases of categories from all Specific Artificial Intelligences for Artificial Research by Application.

As fifth phase, the most important one in order to settle our first human interaction - Artificial Intelligence, is going to be as a result to start experiments, at particular level, about how to construct the first replicas of our human brain, in order to get ready for the first steps in the transcending process, whose last aim is the complete synthesis between human brain and Global Artificial Intelligence up till the sixth phase, and from the seventh phase on the complete synthesis between human mind and the pure reason itself. In further developments, such as the eight phase, one possible development of this direct interaction between human mind and pure reason itself would be our multiple virtual lives in all those multiple parallel universes as a result to the creation of that matrix formed by all possible combination of variations based on the pure equation, in which the matrix of data from the sixth phase has been reduced to a matrix of equations.

The sixth phase is no other thing than the synthesis of the global matrix from the standardized Global Artificial Intelligence but now as factual hemisphere in the matrix in the integrated Global Artificial Intelligence, as long as the unified database of categories in the Unified Application is now the conceptual hemisphere of the matrix in the integrated Global Artificial Intelligence.

The matrix as first stage in the integrated Global Artificial Intelligence provides all the information necessary as to make deductions in the second stage of the integrated Decisional System, so the Artificial Research by Deduction in the Global Artificial Intelligence as a global deductive program, assisted by at least one specific deductive program per sub-factoring level, are going to make global/specific deductions (at this level practically the specific level has become a global level too) to be processed in the third stage across four steps: 1) integrated Modelling System, 2) integrated Decisional System, 3) integrated Application System, 4) integrated Learning System.

Firstly, the deduction made by any global/specific program, is filed by the corresponding program author of this deduction, in the corresponding file in the database of rational hypotheses as the first stage in the integrated Modelling System, where are stored also all those particular rational hypotheses sent by all particular programs. All decisions are going to be checked during this process by the seven rational checks, the first one in the first stage, the rest of them in the second stage, where the integrated Modelling System makes the mathematical models corresponding to every rational hypothesis, at the same time that is carrying out the seven rational comparative checks. Once all possible contradiction in the mathematical model has been fixed, the third stage makes decisions upon mathematical models, using for that purpose, for instance, Impact of the Defect or Effective Distribution, Probability and Deduction (all those rational hypothesis made under Probability and Deduction are in fact at the same time rational hypothesis and decisions), trigonometrical correlations, artificial learning, solving math problems. In fact even those decisions not made directly by Impact of the Defect or Effective Distribution, all the other ones: Probability and Deduction, trigonometry, artificial learning, solving maths problems; their resulting decisions must be assessed using the Impact of the Defect or Effective Distribution in order to label them with some priority level, alike all those ones made directly using Impact of the Defect and Effective Distribution.

Once every decision global/specific or particular, has been labelled with a priority level, the integrated Modelling System files every decision in the right file in the integrated database of decisions as first stage in the integrated Decisional System in accordance with: sub-factoring level (geographical are), encyclopaedic sub-section within that sub-factoring level, and according to priority level within that file corresponding to its sub-section within that sub-factoring level (the stage I will develop in this post). After the first stage, in the second stage of the integrated Decisional System, all decisions are projected, and in the third stage, the projected decisions are transformed into a range of instructions, if passing all the assessments. The instructions are sent to the database of instructions as first stage in the integrated Application System, which will match every instruction with the correct application or robotic device, to be implemented, and after implementation it will send reports about how it was to the Learning System, to analyse further decisions to improve the process.

Throughout all this process, full of different phases, stages, periods , moments, instants, what I will develop in this post is: the second step, in the third stage, in the sixth phase;  the database of decisions in the integrated Decisional System.

And the decisions that are going to be stored in the database, are classifiable in a very similar way than that other classification used in the fifth phase, but now adapted as a possible classification for the integrated database of (global/specific and particular) decisions, having as most important decision to process, what I will call global orders: that type of global/specific decision whose priority level and spatial limits goes beyond a high extreme decision, and implies consequences for all the global model, having as main purpose to keep global harmony.

- First type, high extreme priority decisions, subdivided in: global/specific high extreme decisions, and particular high extreme decisions; and each sub-group sub-sub-grouped in all those different levels of possible high extreme decisions. The main difference between global/specific high extreme decisions and particular extreme decisions, is the fact that by the time a particular high extreme decision has been communicated to the integrated Decisional System, that high extreme particular decision is being implemented directly by the Application System, once that particular high extreme decision has passed a particular quick rational check. While high extreme global decision is a decision that, as soon as it has been stored, is not put into practice after passing the global quick rational check. That means that, if there are two decisions with the same high extreme priority level, but one is particular and the other one is global, the global quick rational check should be first for the particular high extreme priority level, because it is being already implemented, but although the global high extreme decision is not yet on the plan, the global quick rational check for the particular high extreme decision should be done in terms that this must be compared, no only respect to the current decisions on the plan, but additionally with that other global high decision that is going to be checked later, in order that by the time that the global high priority is later on the quick rational check, any possible contradiction with that particular one, would have been adjusted previously, so the quick rational check for the global high priority decision must be comparing this one with the rest on the plan. Once that particular high extreme decision has been assessed (as it has been explained) by the global quick rational check, having found any contradiction with this decision and any other one already on the plan, depending on priority and/or origin (global/specific or particular) must be made the adjustments: always the one to be adjusted in case of contradiction is that one with the lower priority, but in case that two different high extreme decisions have a contradiction and both have the same high extreme priority, then the one to be adjusted is the particular one, adjusting the particular one to the global/specific decision. Any necessary adjustment on any particular decision on the plan, should be communicated to that particular program responsible for this decision to include as soon as possible all the adjustments on the mathematical expression of this decision in its particular database of decisions, in order to make the corresponding projects, and particular adjustments if necessary, looking forward to the immediate transformation of this new adjustments into a new range of instructions to substitute the current ones on the particular Application System.

- Second type, extreme priority decisions, whose priority level is lower than high extreme priority decisions, having two main sub-groups: particular extreme priority decisions and global/specific extreme priority decisions; and each sub-group is sub-grouped into as many sub-groups as sub-categories of different extreme priority decisions could be distinguished. The main difference between particular high extreme decisions and particular extreme decisions, is the fact that particular extreme decisions are not being implemented yet, although having passed the particular quick rational check, by the time they are sent to the integrated database of decisions, waiting for the global authorisation for their implementation. Due to this very important difference, because these particular extreme decisions are not yet implemented yet, waiting for the global authorization, issued by the integrated Decisional System, in case that there are simultaneously two extreme decisions, one global/specific and the other particular, because both of them are under the same circumstances, waiting for global authorization to be implemented, in this case now always the first one to pass the global quick rational check is always the global/specific extreme decision, and secondly the particular extreme decision. And in case that there is any contradiction between two decisions, regardless of their origin, global/specific or particular, it is always the one with lower priority that one to be adjusted in order to avoid the contradiction. But if two decisions have the same priority level, one is specific/global, the other particular, the particular decision is the one to be adjusted to the global/specific.

- Third type, normal decisions, including as a whole particular normal decisions and global/specific normal decisions, as all those decisions, global/specific of particular, neither extreme, routine, nor automatic decisions, to pass the seven global rational adjustments (although particular normal decisions have already passed the particular seven rational adjustments, the seven global adjustments will be necessary to keep the harmony in the plan for the global model). In fact, the seven rational adjustments are going to track all the decisions, regardless of their priority, but only are going to make adjustments in those with lower priority. However, there can be situations in which an adjustment tracking all the decisions looking for contradictions, could find contradictions in other no normal decisions, but in any case, the seven adjustments always will make the adjustments in those ones with lower priority, otherwise, if there is no mathematical solution by any method (Probability and Deduction, trigonometry, artificial learning, solving mathematical problems), the contradiction is considered as full and that decision with the lower priority is off the plan and sent back to the source to be redesigned.

- Fourth type, routine decisions, defined as those with high relative frequency on the historical records, not having in the past any contradiction on the plan or having some frequency of contradiction the frequency is equal to or less than a critical reason. Including in this type as a whole particular routine decisions and global routine decisions, the main difference between particular routine decisions and global routine decisions is the fact that when particular routine decisions arrive in the integrated database of decisions in the integrated Decisional System, these decisions should already being implemented, not needing any other check more, in order to avoid a funnel effect on the integrated Decisional System. Particular routine decisions should be communicated only to the integrated Decisional System, but do not necessarily need to pass the global quick rational check. The global rational check on routine decisions should be only for global routine decisions.

- Fifth type, automatic decisions, including as a whole particular automatic decisions and global automatic decisions, defining as automatic decision any one that having a direct relation with some combination of measurements in some combination of factors, is possible to set up this decisions by artificial learning as automatic decisions, in order that at any time that on the matrix or on the model or on the plan, this combination of measurements/factors is on, automatically the decision is on the plan. All these decisions, particular and global automatic decisions, should not be assessed as they have a high historical record of reliability; they must be projected on the plan directly, in order to avoid the funnel effect on the integrated Decisional System.

- Sixth type, external decisions, all particular decisions (of any priority or frequency) to be implemented by either robotic devices /applications working for the Global Artificial Intelligence (so these decisions must be sent by the integrated Decisional System to the integrated Application System after passing the corresponding global assessments, quick or normal depending on what kind of decision it is) or other third particular program which in that case after passing the corresponding global assessments, if passing, these decisions are sent to that third particular Application System to pass its particular assessment (quick or normal depending on what kind of decision it is), and if passing, to be implemented by the integrated Decisional System of this third particular program.

- Seventh type, global orders, every new global order on the plan is such a kind of global decision of such a kind of priority, whose main purpose is to keep the global harmony across the global model, and as soon that every new global order is issued by the integrated Decisional System must be implemented. The most important global orders are all those whose main purpose is to keep the stability across the global model.  There are at least two types of global orders, depending on which is responsible for their implementation. The first one is that global order to be implemented by robotic devices or applications working directly for the integrated Global Artificial Intelligence, and managed by the integrated Application System, so that as soon as the instructions of a global order arrive in the integrated Application System is applied as quick as possible by the integrated Application System. The second one, that global order to be implemented by particular programs, distinguishing between a global order directly to only one particular program, and that massive global order to be implemented by more than one particular program, having as limit, the total number of particular programs, so there can be some global order whose implementation will demand the collaboration of all particular program working for the integrated Global Artificial Intelligence. A global order on the plan is a mathematical expression of that decision, able to set or reset the global harmony using, for that purpose, all resources or information available.

A global order should only need a very quick global rational check, especially if there are two or more global orders at the same time, because in that case, that global order with lower priority should be adjusted to that global order with higher priority. In any case, the regular seven adjustments, at any time that any global order, or even high extreme or extreme decision in the database, should make the corresponding rational adjustments in any other decision with lower priority, in order to avoid contradictions, securing the completion of those with higher priority.

In order to comply with all types of decisions, in accordance with their respective method of assessment, quick rational check or adjustment, if necessary,  the method in the management of all these decisions in the integrated database in the sixth phase, is as follow:

- Firstly, every decision is filed by its author (particular program or integrated Modelling System) in the corresponding file of this decision in the integrated database of decisions, which is, according to its priority, filing this decision in the corresponding file of its corresponding sub-section in its corresponding sub-factoring level.

- In order to make the corresponding assessment according to priority and origin, if necessary, in the integrated database of decisions as first stage in the integrated Decisional System, all decisions, according to priority level and origin, the decisions are listed on a list, starting with global orders, followed by global or particular: high priority decisions, extreme priority decision; and ordering the rest of particular or global: normal, routine, automatic, external decisions; in accordance with their respective priority. There can be particular or global routine decisions with high level of priority, for instance, in case of an earthquake, or particular or global automatic decisions with high extreme priority level, such as a fire alarm. In these cases, those routine or automatic decisions having great priority, must be ordered in their corresponding positions according to their priority level within the rest of high extreme priority decisions, even though as routine or mechanic decisions they have passed other different method of assessment, in routine decisions if possible only a global or particular quick rational check, depending on its origin, in automatic decisions not having any assessment at all).

- Another list would be necessary, listing all possible global and particular decisions in accordance with their relative frequency, so at the top of the list are the most frequent decisions, regardless of their priority or origin, and at the bottom of this list are those decisions without any past relative frequency. So as to order all the decisions in accordance with their relative frequency in the past, the integrated Decisional System should have checked in its historical records how many times every decision has been on the plan before, in order to order all decisions by order of frequency. This order what is going to make easier is to recognise any possible routine decision, or new routine decision, for instance, if having stored in the database the same decision in the past, but not so frequent as to be considered a routine decision, as many times this decision again on the plan, its frequency on the records is bigger and bigger up to the point in which a normal decision, having an empirical probability over time equal to or greater than a critical reason, even being in the past normal, as long its frequency grows, can be reconsidered as a routine decision. As many decisions are reconsidered as routine decisions, more and more fluid is the integraged Decisional System, avoiding any funnel effect. However, even if a decision is not a routine decision, all decision, not only routine decision, having in the past some frequency, so in the past was on the plan, having as model the single project that was on the plan in the past, this single model is able to be reuse in the current circumstances, saving time in the second stage of the integrated Decisional Sysem, due to the single project of any project in the past, if stored on its records, can be reuse as many times as this decision is again on the plan, but making as many adjustments as necessary in case of new contradictions-

- In any case, along with the list of relative frequency per decision, another similar is possible to be made using as criteria the frequency of contradictions that a decision, having some relative frequency, had in the past, and storing in the historical records not only the original decision but all possible adjustment that this decision could have in the past, at any time that a decision with some relative frequency is on the plan again, having those adjustments stored on its historical records, is possible from the outset, the first assessment, to adjust this decision according to the contradictions observed, in case that this observed contradiction would be again on the plan during the time that this decision is again on the plan.

- In addition to the lists of: priority and origin, relative frequency, frequency of contradictions; another tool really important in the integrated database of decisions in the integrated Decisional System is the organization of logical sets of decisions, grouping the decisions in accordance with: discrete categories of priority an origin, discrete categories of frequency, discrete categories of frequency of contradictions, one category for each sub-factoring level, one category for each sub-section in every sub-factoring level, and another category for each sub-level including all decision belonging to the same sub-section across all different sub-factoring levels. So every decision must be stored simultaneously in its corresponding set according to priority and origin, its corresponding set according to relative frequency, its corresponding set according to frequency of contradictions, if any, its corresponding set according to sub-factoring level, its corresponding set according to its sub-section in its sub-factoring level, its corresponding set according to its sub-section in genera, where all decisions belonging to the same sub-section across different sub-factoring levels are represented.

This last tool, the logical organisation of decisions in logical sets, will allow us to work with large amounts of decisions in order to make faster and easier all the assessments required in the first stage in the integrated Decisional System, working with them using diagrams of Venn.

In general the order to follow for the assessments is in accordance with priority and origin, as it was explained in every one of the seven type of decisions, having in mind the importance of frequency for routine and automatic decisions, and having in mind how important the frequency the contradictions in the past is in order to make from the first assessment faster and easier adjustments at any time that a contradiction is found.

If at any time that a contradiction with a higher priority is in the database of decisions, the first rational adjustment identify what other decision with lower priority is necessary to adjust, identifies as well if this contradiction was observed in the past, and on the historical records there are some models of adjustments made on this contradiction in the past, directly, instead of the calculation of a new mathematical solution for this contradiction, can reuse that solution used in the past if suitable.

Rubén García Pedraza, 7th of October of 2018, London

Reviewed 21 October 2019, Madrid

Reviewed 17 May 2025, London, Leytostone

The second stage in particular applications for particular programs

The second stage in any Artificial Intelligence is the replication stage, in which the Artificial Intelligence is going to imitate human skills. The replication could be at two different levels, at a robotic level and at artificial psychological level. The robotic replications are going to replicate all the human physical skills, while the artificial psychological replications are going to replicate all the human psychological skills.

Because there are two types of replications, robotic and artificial psychological, in the third stage of auto-replication, the auto-replication stage, there are two different subjective auto-replications: robotic subjective auto-replications, and artificial psychological subjective auto-replications. As main difference between subjective and objective auto-replications, is the fact that subjective auto-replications, robotic or artificial psychological, are auto-replications to improve or enhance, in this case, the artificial researcher itself, while objective auto-replications are mainly focused on the improvement in the object of investigation, the reality itself, through improvements in matrix and models, at specific, particular or global level, to make decisions to protect and better the object itself, the reality.

In the third period of consolidation in the fifth phase of collaboration between by Application and by Deduction, when the phases third (standardization) and fourth (unification), are in their respective consolidation periods as well, the main differences between the stage of the application and the stage of replication in a particular application for a particular program itself, is the fact that: 1) the particular integrated matrix, managed by the particular integrated application, as the first stage of application in a particular application for a particular program, imitates the human brain structure distributed in two hemispheres, additionally to the imitation of the human comprehension skills as the second stage of replication within the particular integrated application itself 2) while the second stage of replication in a particular application for a particular program itself imitates human explanation skills.

The way in which the particular integrated matrix is going to imitate the human brain structure is through the organization of the particular integrated matrix in two hemispheres: the conceptual hemisphere based on categories and the factual hemisphere based on factors, including factors as subjects and factors as options.

This double structure is due to the particular integration process between: 1) former particular applications which originally, in the first moment of experimentation in the second period of formation in the phase fifth, came from Specific Artificial Intelligences for Artificial Research by Application, which still work during the first period of coexistence in phases third (standardization) and fourth (unification),   2) former particular programs that some of them were as well originally former Specific Artificial Intelligences for Artificial Research by Deduction, still working in the coexistence period in phases third and fourth.

And when the particular integration is done, the particular integrated application is responsible for the management of the particular integrated matrix, at the same time that as a former particular application, internally has the three stages of application, replication, and auto-replication, inherited from former Specific Artificial Intelligences for Artificial Research by Application and former particular applications, and that is why particular integrated applications in their own second stage are responsible for the artificial comprehension of that particular thing or being, whose information, conceptual and factual, is managed in the particular integrated matrix, using especially the conceptual information to make particular conceptual: schemes, maps, sets, models.

The reason why the particular integrated application manages the particular integrated matrix at the same time that is responsible for the artificial comprehension is because, only that system or person able to understand something, can manage it. Effective management presumes a degree of prior comprehension.

In the selection process of categories from the unified database of categories (third phase) or conceptual hemisphere in the matrix (sixth phase), and the selection process of factors from the  global matrix (third phase) or factual hemisphere of the matrix (sixth phase), along with possible changes in the particular thing or being that demand the selection of the corresponding categories or factors from the global databases or matrix in the third and fourth phases, or the conceptual or factual hemispheres of the matrix in the sixth phase, another reason for this selective process could be the selection of categories and factors in order to fill gaps and blank spaces in the particular conceptual: schemes, maps, sets, and models.

Actually, in the third stage of auto-replication, one way in which the particular integrated application is going to improve the particular matrix is by checking what gaps and blank spaces are in the conceptual: schemes, maps, sets, models; because that means that over these gaps and blank spaces is necessary the inclusión of categories and factors.

Where we do not have concepts, we do not have factors either.

Only it is possible to manage any particular object if previously you are able to comprehend the object itself.

In order to manage the particular integrated matrix, previously, the particular integrated application should comprehend the mechanism of the matrix itself and the object itself. Being this comprehension skills, replication of human skills.

The particular integrated application as responsible for the management of the particular integrated matrix, has then a deep comprehension, something really important in the process of replication of human knowledge because before any rational explanation, it is necessary a conceptual comprehension.

Before modern science was able to explain the reason behind: the day and the night, the lunar cycles, the annual seasons, the growth of the plants, why we are born, why we need food, why some plants have medical effects to cure some diseases, etc., even in ancient times, the first humans comprehended that at any time that you throw a stone or an arrow to the sky, depending on your own force when you throw it and the speed of the wind, sooner or later the stone or the arrow falls to the ground,  they comprehended that every year regularly the annual seasons follow each other, the day happens the night, and vice versa, and the first humans comprehended that they needed food and water to survive.

They could not have been able to explain these facts scientifically because they did not have our modern rational science, and they started explaining these phenomena using fairy tales, myths, legends, and finally, the creation of religion as the first not scientific attempt to explain the because of the causes, and owing to the contradictions between religion and reality, in modern times many scientists, even though with very deep religious beliefs, such as Copernicus or Galileo, started the modern science: the rational science.

The first humans were not able to explain rationally why they needed to drink water or eat every day, but they comprehended that without water or food, they would die. They were not able to explain rationally the function of that red liquid inside their body, the blood, but they comprehended that if you are injured and you do not stop the red liquid coming out your body, you could die, and they comprehended, although not knowing the rational explanation, that at any time they killed an animal, they could get meat to eat and survive.

The first humans, even without the modern rational scientific explanation that we get through modern rational science, comprehended that every time you throw a stone or an arrow into the sky, it falls to the ground sooner or later, and for that reason, even using this primitive comprehension, even not having a rational explanation, they produced the first primitive weapons, to hunt, to protect themselves, or for the very first primitive wars between tribes.

The first humans were not able to explain rationally the day and the night, but they comprehended that every day there is a sunrise and a sunset, in every night the moon has different shapes, and using this comprehension, even not having a rational explanation, they were able to create the first moon´s calendars, and using the first moon´s calendars, even without rational explanation for them, they were able to predict when the time of the rains was coming, and when was the best time to seed or to harvest.

Because they comprehended these facts, even not having a rational explanation, they were able to: they developed tools, early calendars, and strategies for agriculture and survival based on this intuitive understanding. 

What this means is the fact that the sequence of human knowledge is: firstly, we comprehend, and once we have a comprehension of the phenomena, we try to explain.

The first explanations in human history were not rational explanations, but religious explanations, and because of the contradictions between religious explanations and the reality itself, the first modern scientists proposed the first rational explanations, using mathematics as rational knowledge, appearing the first modern scientists such as Copernicus or Galileo, proposing the use of mathematics to explain the universe.

Thanks to this modern rational explanation of the universe, our scientific decisional model is much more advanced than the previous one based on religion in ancient times. For instance, because we know what gravity is, and we have developed a rational technology, we can put into orbit a spaceship or any artificial satellite. Because we can rationally explain the functions of the blood in our body, we can make decisions. For instance, when somebody suffers a haemorrhage, we can stop it and make a transfusion to save his life.

The rational explanation of the world allows us to make better decisions, and develop more advanced technology in order to save lives and make our lives easier and more comfortable.

In general, the sequence of human knowledge is as follows:

- Firstly, in order to know something, we need to comprehend it.

- Secondly, if we have developed a very deep comprehension, we can explain it.

- Thirdly, if we can explain why something happens, we can make better decisions related to this particular matter in order to improve our living conditions.

The sequence of human knowledge could be synthesized in comprehension (thesis, conceptual), explanation (anti-thesis, mathematical), decision (synthesis, praxis: the praxis as a synthesis of concepts, overall ethic concepts, and mathematical factors, for instance, engineering, put altogether into action).

Following this order, the way in which finally particular applications  work for particular programs (as an experiment in order to prepare the integration process of the Unified Application and the Artificial Research by Deduction in the Global Artificial Intelligence) is:

- First stage of the particular applications for the particular programs: the particular integrated matrix is going to imitate the structure of the human brain organized in two hemispheres: conceptual (categories) and factual (factors); and the particular integrated matrix is going to be managed by the particular integrated application responsible too for the artificial comprehension of the particular thing or being making all possible conceptual: schemes, maps, sets, models; related to the particular thing or being to study.

- Second stage of the particular applications for the particular programs: the Particular Deduction Program is going to make a rational hypothesis based on rational relations between factors and sub-factors, at any level of sub-factoring, within the factual hemisphere in the particular integrated matrix, in order to get a rational explanation about the particular object to study.

- Third stage of the particular applications for the particular programs: all the decisions oriented to the improvement and enhancement of the object itself (objective auto-replications) and the subject itself (subjective auto-replications: in robotics or artificial psychology). Objective auto-replications are going to be made through 1) improvements in both hemispheres, conceptual and factual, in the particular integrated matrix, by the inclusion of new categories and/or factors due to new findings and new rational hypotheses, improvements that are going to produce 2) improvements in particular conceptual: schemes, maps, sets, models; as well as 3) improvements in the particular comprehensive model, that later are going to be added to the global comprehensive model, and improvements in Virtual or Actual, Prediction and Evolutionary, Models, being improvements whose final result is to make further decisions to send to the Decisional System in order to protect or better the real object itself in the reality, through robotic devices coordinated by the Application System.

In brief, this process is going to replicate: the first stage is the imitation of the human brain and human comprehension, the second stage is the imitation of human rational explanations, the third stage is the imitation of the way in which we humans make rational decisions. In synthesis, the sequence is first-stage comprehension, second-stage explanation, and third-stage decision.

In particular applications for particular programs, the responsible for each stage are: the particular integrated application for the first stage, the Particular Deduction Program for the second stage, the Modelling System at a particular level for the third stage sending its particular decisions to the Decisional System, and if rational, put then the Application System into practice, being the Learning System at the end responsible for the assessment of the whole process.

The Particular Deduction Program, as the second stage in particular applications for the particular programs, is going to make deductions searching for mathematical relations in any combination of factors across the factual hemisphere in the particular integrated matrix.

The factors in the particular integrated matrix consist of: 1) all those factors that the particular integrated application, as a manager of the particular integrated matrix, has chosen from the global matrix (third phase), or the matrix (sixth phase). Among the reasons for the selection, one of them is to fill the gaps and blank spaces in conceptual: schemes, maps, sets, models; being aware that there are gaps or blank spaces because, are not included yet the correct concepts or factors for these gaps and blank spaces, as well as the selection can be due to changes in the particular thing or being, and 2) all possible factors that any robotic device, working for any remaining Specific Artificial Intelligence (remaining from the first phase), or working for any remaining particular application (remaining from the second period of formation in the fifth phase), or working for any other particular application for any other particular program (emerged after the completion of the consolidation period in the fifth phase),   could set up directly in the global matrix (third phase) or the matrix (sixth phase), being susceptible to be chosen by any particular application for particular program, 3) and all possible factors that those robotic devices working for a particular application for a particular program could set up in the factual hemisphere of its particular integrated matrix as well as the global matrix (third phase) or the matrix (sixth phase).

The way in which Particular Deduction Programs work as the second stage in Particular Applications for Particular Deduction Programs within Artificial Research by Deduction in the Global Artificial Intelligence, is in the same way that Particular Deduction programs work in the second period of formation in the fifth phase, when the Particular Deduction Programs were formed only as Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence, before the particular integration process between these programs and the particular application in the consolidation period in the fifth phase.

Before the third period in the fifth phase, many Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence, were only the way in which former Specific Artificial Intelligences for Artificial Research by Deduction, not having being absorbed by the Artificial Research by Deduction in the Global Artificial Intelligence itself, became Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence, as it was explained in the last post “The first stage in particular applications for particular programs”.

So the way in which Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence work is in the same way that the former Specific Artificial Intelligences for Artificial Research by Deduction work, with the difference that the particular matrix has not so strong academic limits and not so strong spatial limits, due to it can have factors from any synthetic science, discipline, activity, in any location within the spatial limits of the global matrix. Among other differences explained in the last post “The first stage in particular applications for particular programs”.

And having not so strong academic and spatial limits, the way in which the Particular Deduction Programs work in the Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence, is similar to the former Specific Artificial Intelligences for Artificial Research by Deduction: searching for mathematical relations in any combination of factors, at any level of sub-factoring. The same way to work for the Particular Deduction Programs as second stage now in the Particular Applications for Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence, the particular applications for particular programs.

The Particular Deduction Programs, now as the second stage of replication in the particular applications for the particular programs, are going to make deductions searching in the factual hemisphere in the particular integrated matrix, any possible mathematical relation in any possible combination of factors, at any level of sub-factoring.

The deduction process is as follows:

- The Particular Deduction Program tracks the factual hemisphere in the particular integrated matrix, looking for any mathematical relation in any combination of factors at any level of sub-factoring.

- Every time a Particular Deduction program finds out a possible mathematical relation in any possible combination of factors, at any level of sub-factoring, this relation in this combination is considered an empirical hypothesis.

- The Particular Deduction Program gathers a sample of data for each factor involved in the empirical hypothesis. The sample could be from the past or the future. If from the past, the Particular Deduction Program then gathers from the factual hemisphere in the particular integrated matrix some flow of data from the past in each factor involved (the limits about how old the data should be, could be programmable according to different types of possible situations in which are necessary to gather data from the past). If in the future, after waiting some time from the very moment in which the empirical hypothesis was made, the Particular Deduction Program takes as a sample of the flow of data for every factor the flow of data from the moment in which the empirical hypothesis was made until the moment in which it stops waiting (how long does it would wait, is programmable according to different possible situations as well).

- Having the Particular Deduction Program samples of data, the Particular Deduction Program contrasts rationally the empirical hypothesis, and if rational, it becomes a rational hypothesis to be included in the rational truth. The way to carry out the rational contrastation could be made using statistical methods, probabilistic methods, or any other method according to the nature of the hypothesis, but in any case, the contrastation must always be mathematical, rational. The rational truth is all the set of rational hypotheses gathered in the database of rational hypotheses. The rational global truth is the global database or rational hypothesis. The particular rational truth is the particular database of rational hypotheses. However, the whole particular rational  truth must be integrated as well in the global rational truth, which means that all the rational hypotheses in the particular database of rational hypotheses must be included as well in the global database of rational hypotheses.

The rest of the process of how to make single virtual models, particular comprehensive virtual models, and how to integrate them in the global comprehensive virtual model, the global model, corresponds to the third stage of auto-replication. Although the protective decisions, after the application of the Impact of the Defect, could be considerable part of the second stage, and bettering decisions strictly related to auto-improvements. However, all types of decisions, as decisions, are going to be developed finally as a third stage: the stage of auto-replication is going to be considered at the end as a decision stage itself.

The mathematic relations in any combination of factors, as it was explained in the post “Replication processes in the Specific Artificial Intelligence for Artificial Research by Deduction”, could be at least: stochastic, patterns (including patterns in a group of factors or individual patterns in every individual factor), cryptographic, and in the Second Method of Impossible Probability relations of equal opportunities or bias, positive or negative. In addition to any other mathematical method of analysis, that from other mathematical disciplines could be suitable to add to have a much deeper analysis of the phenomena. As I have said in another post, these posts about Global Artificial Intelligence from the point of view of Impossible Probability are only a humble contribution in order to create the first model of Global Artificial Intelligence, whose final result is going to integrate contributions from different mathematical disciplines, not only from the statistical and probabilistic perspective, and is going to include different traditions and philosophies, depending on the culture of all those first countries involved in its first model.

Among the stochastic relations, in Impossible Probability are included: relations between probable causes and effects, possible directly proportional positive correlations, possible directly proportional negative correlations, and possible inversely proportional correlations.

The deductions of mathematical relations in any combination of factors, as it was explained in the post “The standardization process in the second stage”, are deductions which in turn can be classified as: deductions of mathematical relations in combinations of factors, including only factors as subjects, deductions of mathematical relations in combinations of factors including factors as subjects and factors as options, and deductions of mathematical relations in combinations of factors including only factors as options.

And in addition to the possible classification of factors in factors as subjects or as options, is necessary the distinction between: constant factors, and variable factors; distinguishing: factors as independent variables, and factors as dependent variables.

The classification of factors in subjects or options in Impossible Probability, depends on the way in which the factors are mathematically measured. If a factor using a scale of measurement is measured in direct punctuations then the factor works as a subject. If a factor is measured by counting its frequency then the factor works as an option.

The classification of factors in: constant, or variable, dependent or independent; depends on their behaviour, in Impossible Probability is considered the tendency. If the behaviour behaves keeping constant all the measurements regardless of any other circumstance, the factor is a constant. If not the factor is variable. If the factor is variable could be dependent or independent, something really important in mathematical relations of probable causes and effects, due to the independent variables work as probable causes, and the dependent variables work as probable effects.

If the behaviour of a factor depends on some circumstances, the factor is a dependent variable, so it is a probable effect of such circumstances. If the variable behaviour of one factor produces changes in other factors, the first factor is the independent variable, probable cause, for the other factors, probable effects, as dependent variables of the first factor as probable cause.

But, at the same time, the variable behaviour of the first factor, as it has not a constant behaviour, because it has not constant measurements, as a variable, even independent for those other following variables, the first independent variable could be in turn a dependent variable, possible effect, depending on the changes of previous factors, as probable causes, in a long chain of factors, in which every single factor could be dependent (effect) on the previous factor as independent (cause), at the same time that this single factor could be independent (cause) for the following factors (effects) in the chain.

Dialectically, one factor could be at the same time cause and effect, effect in relation to the previous factors, cause in relation to the following factors. In the end, one more time, we see how the opposites are dialectically identical: cause and effect are the same. The same reality could be explained as a chain of probable causes, or as a chain of probable effects, or as a chain of probable causes and effects, in which every probable cause itself is at the same time a probable effect itself.

One of the objectives of the particular models at a particular level is to draw how this chain of causes and effects works, and one of the objectives of the global model at a global level is to draw the chain of factors as causes and effects at a global level.

Because there are at least two classifications of factors, according to their measurement (subjects, options), and according to their behaviour (constant or variable, dependent or independent, in mathematical relations related to probable cause and effect), the synthesis of both classification in only one is:

- Constant factors as subjects, keeping constant their direct punctuation.

- Constant factors as options, keeping constant their frequency.

- Independent factors as subjects whose changes in their direct punctuations can produce changes in other factors as subjects (changes in their direct punctuations) or as options (changes in their frequency).

- Independent factors as options, whose changes in their frequency can produce changes in other factors as subjects (changes in their direct punctuations) or as options (changes in their frequency)

- Dependent factors as subjects whose changes in their direct punctuations are due to changes in the direct punctuation of other factors as independent factors as subjects, or due to changes in the frequency of other factors as independent factors as options.

- Dependent factors as options whose changes in their frequency are due to changes in the direct punctuation of other factors as independent factors as subjects, or due to changes in the frequency of other factors as independent factors as options.

Tracking the factual hemisphere in the particular integrated matrix, the Particular Deduction Program should be able to find any mathematical relation in any combination of factors, including as possible mathematical relations all those relations between independent and dependent variables as probable causes and effects, in order to draw later on the third stage by the Modelling System at a particular level the possible chain of causes and effects in the particular thing or being, that later are going to include in the global model.

The way in which the Particular Deduction Program is going to track the flow in the factual hemisphere in the particular integrated matrix to look for relations between factors, depends on how it is going to be organised in the particular matrix, if as a collection of single factors or composed factors.

If the factual hemisphere is organised as a collection of single factors, the flow to track then is the flow of data, tracking the flow of data coming up from all single factors. If the factual hemisphere is organised as a collection of composed factors, the flow to track then is the flow of packages of information, tracking any possible mathematical relation in any combination of factors at any level of sub-factoring, in every package of information, and tracking any possible mathematical relation between factors from different level of sub-factoring and from different packages of information from different composed factors.

Once any possible mathematical relation (stochastic, a pattern even at an individual level not only between different factors, cryptographic, equal opportunities or bias, positive or negative) is found between any combination of factors, regardless of their level of sub-factoring and original composed factors, the mathematical relation is considered as an empirical hypothesis to contrast, and if rational, as a rational hypothesis belongs to the rational truth, the database of rational hypothesis, at particular and global level, in order to be modelled by the Modelling System, in order to make decisions after the application of the Impact of the Defect and the Effective Distribution ( the name in which finally was published in Introducción a la Probabilidad Imposible, estadística a la probabilidad o probabilidad estadística, the Hierarchical Organization).

The way in which the flow of packages of information in the factual hemisphere in the particular integrated matrix could be tracked by the Particular Deduction Program is like when you try to search for a file, or even for a simple word, in your computer, your computer looks for this file, or this simple word, across all the folders, or sub-folders in any folder, or sub-sub-folder in any sub-folder within any folder, or any other level of sub-folder.

In any case, the way in which the Particular Deduction Program is going to carry out the research process in the second stage as an explanation stage, making deductions from the factual hemisphere in the particular integrated matrix. As well as the way in which the particular integrated application is going to carry out all the processes to develop a deep artificial comprehension through the creation of particular conceptual: schemes, maps, sets, models; at the same time that the particular integrated application is going to manage the particular integrated matrix structured in two hemispheres, conceptual and factual. In addition to the experiments about how the Modelling System at a particular level can work, in order to make single or comprehensive virtual models, and Virtual and Actual, Prediction Evolutionary, Models, in order to make decisions after the application of the Impact of the Defect to make protective decisions, and the Effective Distribution to make bettering decisions, in order to improve the object, if the Decisional System accepts the decisions as rational, putting them into practice by the Application System, is as a whole a long process of experimentation at particular level, whose most important result is the application of the most successful results of this experimentation into the sixth phase: the integration process at global level to create the final model of Global Artificial Intelligence.

Along this long process of collaboration and competition among global stakeholders, something that is really important to be aware of, is the fact that the scientific policy that must rule the final model of Global Artificial Intelligence, must be based on values such as democracy, freedom, and human rights, in order to be the most important way to protect the global peace.

For that reason, it is necessary that any remaining Specific Artificial Intelligence, even those ones based on artificial learning, that would not have been absorbed by the Global Artificial Intelligence, or would not have become particular applications or particular programs or particular applications for particular programs working for the Global Artificial Intelligence, in any case, any remaining Specific Artificial Intelligence, included those ones based on artificial learning, hould be aligned with the scientific policy of the Global Artificial Intelligence, made by those international agencies responsible for the global peace, in order to protect the humanity against any use of any Specific Artificial Intelligence, including those ones based on artificial learning, that can put at risk the global peace.

Rubén García Pedraza, 12 th of May of 2018
Reviewed 18 August 2019 Madrid

Reviewed 10 August 2023 Madrid

Reviewed 9 May 2025 London, Leytostone

The standardization process in the second stage

In the standardisation process, all databases from all kinds of agencies and institutions and specific matrices from Specific Artificial Intelligence for Artificial Research by Deduction are going to be shared in the same database, as an Application for the foundation of the Artificial Research by Deduction in the Global Artificial Intelligence.

Firstly, this foundational database is going to be a gigantic database, that later, through a long process of standardization, is going to be moulded in order to transform all information into the same format, following a homogeneous format of factor: 1) defined in quantitative terms, 2) defined whether it is a single factor able to provide a flow of data or a composed factor able to provide a flow of packages of information, which in turn is going to contain sub-factors at a different level of sub-factoring, 2) including, either as a single factor, composed factors, or sub-factors, factors as subjects and factors as options.

As a result of the standardisation process, the product is going to be the global matrix. Due to the enormous dimensions of this first global matrix, in order to track it in the second stage of replication, in order to save time and energy and gain efficiency and velocity, one solution could be the distinction of two different periods in the standardization process: the period of coexistence and the last and final period of consolidation of the global matrix.

The period of coexistence means that, while the global matrix is still tested and lots of experiments are going to be carried out on the global matrix about how to know which is the best way to track the global matrix to make faster deductions and decisions, during this time of full experimentation over the global matrix is possible to keep on working the original Specific Artificial Intelligences for Artificial Research by Deduction, in all synthetic sciences, disciplines, and activities where they have been originally created, in order to save time and energy, because while the Specific Artificial Intelligences for Artificial Research by Deduction still work on specific deductions, in that case, the Artificial Research by Deduction in the Global Artificial Intelligence can be completely focused only on global deductions.

During this period of coexistence, along with the global matrix as a foundational stone for the creation of Artificial Research by Deduction in the Global Artificial Intelligence, making the first global deductions, and while the former Specific Artificial Intelligences for Artificial Research by Deduction still work on their specific synthetic science, discipline, activity, at the same time it could be created the first Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence.

The Particular Deduction Programs are not going to be intelligence itself because they are not going to have their own application; they are going to work directly, choosing from the global matrix all those factors that they need to make their own deductions in that particular thing or for what they have been designed.

For instance, it could be created particular programs for particular industries, particular factories, particular means of transport (from the particular program for the airport of San Francisco to the particular program for every single train from London to Paris), particular programs for schools, hospitals, a particular program for the court, particular programs for particular security systems, or surveillance systems. Or even a particular program for a human being, or particular programs for animals, like whales, animals in danger of extinction, fish banks, herbs, migratory birds…

In the same way that today a personal assistant in your personal computer is able to make suggestions of possible decisions based on results on your Social Media or searches on browsers on the internet, a particular program should have access to the global matrix and choose those factors to make deductions on the particular thing or being for what it has been created.

In order that such a system of particular programs can work, it is going to be absolutely necessary that the matrix has access to absolutely all possible information without restriction.

Due to the moral dimensions that such a project has, the psychological paradigm for the artificial psychology within the Global Artificial Intelligence, must be a liberal paradigm, in order that the Global Artificial Intelligence must respect any personal behaviour or thought if there is no contradiction between any behaviour or thought and the law, and does not put at risk anything, or is not harmful for anyone without permission. In case of harmful behaviour to another person, but with permission, such as abortion, euthanasia or some sexual practice, should be permitted.

This kind of technology is only going to be possible through a liberal paradigm in artificial psychology within the Global Artificial Intelligence, making possible the permanent protection of democracy, freedom, human rights, and perpetual peace. Otherwise, if the Global Artificial Intelligence does not follow a liberal paradigm, we could be dragged to terrible and unforeseen consequences. The construction of such technology is full of opportunities, but with many risks.

Coming back to the benefits of the construction of Particular Deduction Programs within the Artificial Research by Deduction in the Global Artificial Intelligence, the main benefit is the fact that while the particular programs work on particular deductions in order to make particular decisions for particular things or beings, the Artificial Research by Deduction in the Global Artificial Intelligence can be focused only on global deductions in order to make global decisions.

As long as the standardization process goes on, and Artificial Research by Deduction in the Global Artificial Intelligence is able to improve its skills in making global decisions, at the same time that many more Particular Deduction Programs are created, there is going to be a moment in which gradually the Specific Artificial Intelligences for Artificial Research by Deduction are going to be completely absorbed by the Artificial Research by Deduction in the Global Artificial Intelligence, or are going to be transformed into Particular Deduction Programs.

When the Specific Artificial Intelligences for Artificial Research by Deduction disappear, the standardisation process is in the second period of consolidation of the global matrix.

In the first period of coexistence, there are three kinds of deductions: global deductions, specific deductions, and particular deductions.

As long as the process goes from the first period to the second period, the specific deductions will disappear. Some of them are transformed into global deductions, others into particular deductions.

At the end of the standardisation process, the last and final period, the consolidation of the global matrix, there are only two possible deductions: global deductions and particular deductions.

From the global deductions are going to be generated global decisions, and from the particular deductions are going to be generated particular decisions.

These global or particular decisions could be:

- Global protective descriptive research decisions

- Global bettering descriptive research decisions

- Particular protective descriptive research decisions

- Particular bettering descriptive research decisions

The reason why it is very important to clarify every kind of decision is that later on in the Decisional System is going to be very important to know the origin and level of every decision. But about how it is going to work, the production of decisions in the standardisation process is much more related to the third stage than the second stage. I only mention the relation between deduction and decision to be aware of how the deduction process affects the decisional process, but the decisional process, along with the standardisation process, is going to be more developed in the next post.

What is really important in the second stage of replication during the standardization process is how to make deductions, globally or particularly, minding that within the global matrix, either it is organized by single factors or composed factors, among the factors or sub-factors are going to be included factors as subject and factors as options.

In any kind of deduction, global or particular, keeping in mind that in the global matrix, there are factors as subjects and as options, there are at least three different ways to make deductions.

- Mathematical relations in combinations of only factors as subjects.

- Mathematical relations between one factor as a subject, or a combination of factors as subjects, and one factor as an option, or a combination of factors as options. Or vice versa, mathematical relations between one factor as an option, or a combination of factors as options, and one factor as a subject, or a combination of factors as subjects.

- Mathematical relations in combinations of only factors as options

The possible mathematical relations were explained in the post “Replication processes in the Specific Artificial Intelligence for Artificial Research by Deduction”; the only difference is the fact that instead of being a deduction process applied on a specific matrix, it is now on a global matrix, but the process itself is the same. Mathematical relations are going to be understood: stochastic mathematical relations, mathematical patterns, possible cryptographic relations, relations of equal opportunities or bias, positive or negative. 

And more precisely, stochastic relations are understood:

- Probable cause and effect, the mathematical observation that after some changes in a factor or combination of factors, there are changes in other factors or a combination of factors.

- Possible directly positive proportional correlations, when parallelly one factor or a set of factors increases, another factor or set of factors grows at the same time.

- Possible directly negative proportional correlations, when parallelly one factor or a set of factors decreases, another factor or set of factors decreases at the same time

- Possible inversely proportional correlations, when one factor or a set of factors increases, at the same time another factor or set of factors decreases, or vice versa.

For mathematical pattern is understood when a factor itself or a set of factors has changed following a special rule that could be deduced. This rule could be a repetition in circles or spirals, or an increase or decrease rule, or any other one that makes a behaviour completely predictable.

The reason why I think that cryptography could be advisable to include in those mathematical methods to analyse permanently the global matrix is due to the similarities that some mathematical patterns have with some cryptographic methods.

And finally, the inclusion of analysis methods from the Second Method of Impossible Probability, such as studies in equal opportunities or bias, positive or negative, in order to know what factors, as subjects and as options, in the global matrix show a behaviour explainable by equal opportunities, so it could be a random behaviour (but it must be checked), and what factors show a biased behaviour, positive or negative. Any biased behaviour is not random behaviour,  so there must be some cause behind to research.

At any time, by the Second Method of Impossible Probability, any behaviour is found explainable by equal opportunities, so it could be random. Still, it must be checked, or any biased behaviour, should undergo further analysis to explain this behaviour and the possible cause that produces this behaviour.

In essence, for mathematical relations are going to be understood: stochastic relations (cause and effect, direct positive or negative proportional correlations, inversely proportional correlations), patterns, cryptographic correlations, and from the Second Method, relations of equal opportunities or bias, positive or negative; mathematical relations in any combination of factors that are going to be treated as possible deductions to form empirical hypothesis, in order to contrast the empirical hypothesis rationally, and if rational, now as a rational hypothesis belonging to the rational truth, the formation of a single virtual model to include in the comprehensive virtual model.

This process of deduction, from the very beginning makes mathematical relations between combinations of factors, up to the formation of single virtual models to include in the comprehensive virtual model, must integrate factors as subjects and factors as options, even at any level of sub-factoring in case that the global matrix is built following the format of composed factors (whose flow is a flow of packages of information) rather than single factors.

In order to integrate deductions from factors as options and/or subjects, it is necessary to know how it is possible to set up deductions having different kinds of factors, either as subjects or as options.

Due to the speciality of Impossible Probability is probability and statistic I will only give a brief explanation about how to make these deductions in this field, so I will give a brief explanation of stochastic relations and relations within the Second Method.

Firstly, deductions from combinations of only factors as subjects, defining a factor as a subject like that one whose flow of data is a flow of direct punctuations.

- A probable relation of cause and effect between factors as subjects is when, after some changes in the flow of direct punctuation from a factor as a subject or a set of factors as subjects, there are other changes in the flow of direct punctuations of another factor as a subject or other factors as subjects.

- A possible direct positive proportional correlation between factors as subjects, is when parallelly one factor as a subject or a set of factors as subjects has an increment in the flow of direct punctuations. At the same time, there is an increment in the flow of direct punctuations in another factor as a subject or a set of factors as subjects.

- A possible direct negative proportional correlation between factors as subjects, is when parallelly, one factor as a subject or a set of factors as subjects has a decrease in the flow of direct punctuations. At the same time, there is a decrease in the flow of direct punctuations in another factor as a subject or a set of factors as subjects.

- A possible inversely proportional correlation between factors as subjects, is when one factor as subject or set of factors as subjects have an increment in the flow of direct punctuations, while at the same time, another factor as subject or set of factors as subjects have a decrease in their flow of direct punctuations, or vice versa.

- A possible relation of equal opportunities between factors as subjects is when all factors as subjects have exactly the same flow of direct punctuations, so if it is calculated the flow of empirical probabilities is calculated, all of them have, within a margin of error, the same value equal to the theoretical probability.

- A possible biased behaviour in factors as subjects is when the flow of empirical probabilities, out of a margin of error, is not equal to the theoretical probability. If the empirical probability is over a margin of error, the theoretical probability, then the empirical probability is positively biased, and if it is below a margin of error, the theoretical probability, then it is negatively biased.

Secondly, deductions from combinations of factors as subjects and factors as options, defining a factor as an option, like that one whose flow of data is a flow of frequencies.

- A probable relation of cause and effect between factors as subjects and factors as options, is when after some changes in the flow of direct punctuations from a factor as subject or a set of factors as subjects, there are other changes in the flow of frequencies of other factor as option or other factors as options. And vice versa, when after some changes in the flow of frequency from a factor as an option or a set of factors as options, there are other changes in the flow of direct punctuations of another factor as a subject or other factors as subjects.

- A possible direct positive proportional correlation between factors as subjects and factors as options is when parallelly, one factor as a subject or a set of factors as subjects has an increment in the flow of direct punctuations. At the same time, there is an increment in the flow of frequencies in another factor as an option or a set of factors as options. And vice versa, parallelly, one factor as an option or a set of factors as options have an increment in the flow of frequencies. At the same time, there is an increment in the flow of direct punctuations in another factor as a subject or a set of factors as subjects.

- A possible direct negative proportional correlation between factors as subjects and factors as options is when parallelly, one factor as a subject or a set of factors as subjects has a decrease in the flow of direct punctuations. At the same time, there is a decrease in the flow of frequencies in another factor, as an option or a set of factors as options. And vice versa, parallel, one factor as an option or a set of factors as an option has a decrease in the flow of frequencies. At the same time, there is a decrease in the flow of direct punctuations in another factor as a subject or a set of factors as subjects. 

- A possible inversely proportional correlation between factors as subjects and factors as options, is when one factor as subject or set of factors as subjects have an increment in the flow of direct punctuations, while at the same time, another factor as option or set of factors as options have a decrease in their flow of frequencies. And vice versa, when one factor as an option or a set of factors as options has an increment in the flow of frequencies, while at the same time, another factor as a subject or set of factors as subjects has a decrease in their direct punctuations.

- A possible relation of equal opportunities between factors as subjects and factors as options is when all factors as subjects  have exactly the same flow of direct punctuations, at the same time that all factors as options have the same flow of frequencies. So if it is calculated the flow of empirical probabilities of factors as subjects, equal to every direct punctuation divided by the total of direct punctuations, all the empirical probabilities of factors as subjects have, within a margin of error, the same value, equal to the theoretical probability for subjects, one divided by the total number of subjects. And if it is calculated the flow of empirical probabilities of factors as options, equal to every frequency divided by the total of frequencies, all the empirical probabilities of factors as options have, within a margin of error, the same value, equal to the theoretical probability for options, one divided by the total number of options. If a set of factors as subjects and a set of factors as options, within their corresponding theoretical probability, are within a margin of error close to their theoretical probability, is possible to make a deduction that, within a margin of error, the behaviour of those factors is explainable by random behaviour, but another possible scenery is the possibility that sometimes the behaviour of equal opportunities could be produced by other factor behind that must he found out.

- A possible biased behaviour in factors as subjects and as options is when the flow of empirical probabilities is, out of a margin of error, not equal to the corresponding theoretical probability. If the empirical probability is over a margin of error, the corresponding theoretical probability, then the empirical probability is positively biased, and if it is below a margin of error, the corresponding theoretical probability is negatively biased. For factors as subjects, the corresponding theoretical probability is one divided by the total number of subjects. For factors as options, the corresponding theoretical probability is one divided by the total number of factors. If a positively or negatively biased subject or subjects are related to a biased option or options, or vice versa, possible deductions could be made. If this behaviour is not casual, and there is something behind it, it could be studied by taking samples from the global matrix, as a deduction to be transformed into an empirical hypothesis to contrast, and if rational, then to proceed to the formation of single virtual models to include in the comprehensive virtual model.

Thirdly, deductions from combinations of factors as options.

- A probable relation of cause and effect between factors as options is when, after some changes in the flow of frequencies from a factor as an option or a set of factors as options, there are other changes in the flow of frequencies of other factor as option or other factors as options.

- A possible direct positive proportional correlation between factors as options, is when parallely one factor as an option or a set of factors as options have an increment in the flow of frequencies, at the same time, there is an increment in the flow of frequencies in another factor as an option or a set of factors as options.

- A possible direct negative proportional correlation between factors as options, is when parallely one factor as an option or a set of factors as options have a decrease in the flow of frequencies, at the same time there is a decrease in the flow of frequencies in another factor as an option or a set of factors as options.

- A possible inversely proportional correlation between factors as options, is when one factor as an option or set of factors as options have an increment in the flow of frequencies, while at the same time, another factor as an option or set of factors as options have a decrease in their flow of frequencies, or vice versa.

- A possible relation of equal opportunities between factors as options is when all factors as options have exactly the same flow of frequencies, so if it is calculated the flow of empirical probabilities is calculated, all of them have, within a margin of error, the same value equal to the theoretical probability.

- A possible biased behaviour in factors as options is when the flow of empirical probabilities is, out of a margin of error, not equal to the theoretical probability. If the empirical probability is over a margin of error, the theoretical probability, then the empirical probability is positively biased, and if it is below a margin of error, the theoretical probability, then it is negatively biased.

Finally, another way in which deductions from factors as options within the global matrix can be made is by considering these options as categories defined in quantitative terms.

If within the global matrix, there are factors as options, for instance, related to diseases, these factors as options about diseases not only are going to be really useful to study their frequency in a country, continent or the whole planet, and how they are spreading out around the country, the continent, or the world, studying their behaviour, making deductions about how their behaviour is related to other factors as options or as subjects, and in case of diseases produced by virus or bacteria, studying their behaviour and how is associated to other factors, either as subjects or as options. The integration of a list of possible diseases within the global matrix working as factors as options (something really useful for the collaboration process between by Deduction and by Application), is going to give the possibility that, for instance, a personal program ( a Personal Particular Deduction Program within the Artificial Research by Deduction in the Global Artificial Research, that one able to make deductions for a particular person in this case), combining the biostatistics of a particular person, and matching the biostatistics with the list of possible diseases permanently, at the least change in any factor in the biostatistics of that particular person which can coincide with any disease included as an option in the global matrix, automatically the personal program can make deductions about what diseases this person could get, in order to make further decisions.

If for an earthquake or a volcano, is necessary that the geological temperature has to grow up to certain point, having a definition about the thermic conditions in which this geological phenomenon happens, at the least sign of the increment of the geological temperature in any place, as long as the temperature is closer to produce some geological phenomenon, a particular geological program ( a geological Particular Deduction Program within the Artificial Research by Deduction in the Global Artificial Research, that one able to make deductions in geology) can make deductions about, calculating the velocity in which the temperature is increasing, and any other factor in the area, how long is going to take the apparition of a geological phenomenon, in order to make further decisions.

Once the global matrix is able to integrate absolutely all possible information, in a country, a continent, the planet, or even the universe, particular programs having access to the information in the global matrix can make deductions in order to make particular decisions, at the same time that the Artificial Research by Deduction can make global deductions to make global decisions.

At the beginning the construction of Specific Artificial Intelligences for Artificial Research by Deduction are going to be an experiment in order to prepare the future construction of the global matrix, but once the global matrix is built then many Specific Artificial Intelligences for Artificial Research by Deduction are going to be completely absorbed by the Artificial Research by Deduction working on the global matrix as a system belonging to the Global Artificial Intelligence (as a system of systems), while others Specific Artificial Intelligences for Artificial Research by Deduction are going to be transformed into particular programs (Particular Deduction Program within the Artificial Research by Deduction in the Global Artificial Research), having the benefit as programs that, much more than a specific matrix, they could have access to the whole global matrix, choosing directly from the global matrix those factors, as options or as subjects, that they will need for the deductions in their particular thing or being, making particular deductions to produce particular decisions.

In this way, what these particular programs are going to prepare is practically the integration process, in which not only some Specific Artificial Intelligences for Artificial Research by Deduction can become particular programs, because some Specific Artificial Intelligences for Artificial Research by Application can become particular programs as well, having access, by the time the integration is finished, to the definitive matrix, the matrix, where all possible categories must be integrated as options, along with the rest of factors, as subjects or as options, already included during the formation of the global matrix.

If, from the outset, the collaboration process between by Application and by Deduction works, the integration process is going to be a mere formality, because, practically, in the global matrix, are going to be included as factors as options all possible categories. The process in which at the end only is going to merge absolutely all possible information about what is happening in the world, or why not, in the universe, in only one matrix, is something that, if Artificial Intelligence goes on evolving towards the absolute knowledge, without restrictions, this process is going to be like a natural evolution process.

This contribution to Impossible Probability is intended as a conceptual and philosophical framework. While other approaches may emerge, I hope this work can offer a useful foundation for ongoing research in Global Artificial Intelligence.

Rubén García Pedraza, 14th of April of 2018, London

Reviewed 14 August 2019 Madrid

Reviewed 10 August 2023 Madrid

Reviewed 4 May 2025, London, Leytostone

imposiblenever@gmail.com