Auto-replication process in the Specific Artificial Intelligence for Artificial Research by Deduction

Auto-replication process in the Specific Artificial Intelligence for Artificial Research by Deduction is the mechanism of this kind of artificial psychology to auto-improve or auto-enhance itself by itself, without human intervention, all its components, from robotics to artificial psychology.

Some ways in which any Artificial Intelligence, Global or Specific, can auto-replicate, were developed in my post “Auto-replicationin Artificial Research by Application”, most of them, except those about how to auto-improve the categories in the database, are valid for any other Artificial Intelligence, Global or Specific, such as the auto-enhance of any Artificial Intelligence using virtual-stores, or other mechanisms through inter-net, intra-nets available only for Artificial Intelligences, Global or Specific, or any other virtual-net, where Artificial Intelligences, Global or Specific, can find advancements to apply on themselves by themselves, advancements  made by any Artificial Intelligence, Global or Specific, and shared within the virtual-net to be used by any other one, or advancements developed by Specific Artificial Intelligence for Artificial Engineering, which is going to be developed in Impossible Probability in the future designing at least two models: the Artificial Designer of Intelligence, and the Intelligence Robotic Mechanic, whose main ideas  where exposed on this blog in “Specific Artificial Intelligence forArtificial Research by Deduction”.

Another way to auto-enhance itself by itself any Artificial Intelligence, Global or Specific, and about what I had written in the post “Auto-replication in Artificial Research by Application”, is the auto-enhancement of the memory through memory release (deleting information not useful any longer), information condensation (using the shortest mathematical expression for any information), and the increase of memory through quantum computing or Artificial Genetics, by the replication of trillions of trillions of  DNA molecules for the storage of artificial information instead of biological information as well as the possibility of replication of trillions of trillions of human neurons  for the storage of artificial information. Or any other method in the future.

All these ways of auto-improvement or auto-enhancement are equally valid for any Artificial Intelligence, Global or Specific.

The very specific way of auto-replication in a Specific Artificial intelligence for Artificial Research by Application, is how to be able to improve its own database, by those empirical hypotheses, which after not being found rational, are chosen as new categories in the database, being this process itself an auto-replication process.

But, in Specific Artificial Intelligence for Artificial Research by Deduction, what is going to auto-improve itself after rational contrastation, is its own comprehensive virtual model, integrated by all single virtual models, from those empirical hypotheses accepted as rational, after the rational contrastation.

The way in which the Specific Artificial Intelligence for Artificial Research by Deduction is built is in general terms: 1) application, database constructed as a matrix where every factor to include is defined in quantitative terms, integrating all factors in the specific subject to study, regardless of how the measurements are taken, in direct punctuations or frequencies, 2) replication, robotic devices fill every file in the matrix adding the permanent flow of data from every factor, tracking, the database, permanently the Artificial Intelligence looking for any mathematical relation (stochastic, pattern, cryptographic, equal opportunities or bias, either positive or negative) in every possible combination of factors (or even at an individual level), relation taken as an empirical hypothesis for the rational contrastation. In those empirical hypotheses accepted as rational, the Artificial Intelligence makes a single virtual model, that later integrates into the comprehensive virtual model, which includes all the rational hypotheses accepted by this Artificial Intelligence, 3) autoreplication, every time the comprehensive virtual model is improved adding new single virtual models, from new rational hypothesis, the comprehensive virtual model experiments an auto-replication, in the sense that by itself is able to auto-improve itself.

The way in which these three stages within the Artificial Research by Deduction are sequenced and must work in any Artificial Intelligence, Specific or Global, must be the same. The only differences are in the algorithms that they may use, and the way in which finally United States, Europe, Russia, or China, introduce different methods of mathematical analysis depending on their own mathematical traditions and philosophy.

In fact, as I have said in the last post, “Replication process in the Specific Artificial Intelligence by deduction”, what I am doing in this blog through my personal contribution to the construction of the first Global Artificial Intelligence is only my personal contribution from the point of view of Impossible Probability, but any other analytical method that could be incorporated in the permanent tracking of the matrix, specific or finally the global matrix, is welcome. The final objective is to have ready the first Global Artificial Intelligence as soon as possible.

The reason why it does not matter what kind of mathematical analysis method is applied to the Artificial Research by Deduction, Specific or Global, is because all the experiments in Specific Artificial Intelligence are not another thing than the preparation of the future construction of the Global Artificial Intelligence. By time through experimentation in Specific Artificial Intelligence, this technology is ready. Its implementation at a global level will be practically to put into something bigger the same mathematic methods that work perfectly in something smaller. But the adjective bigger or smaller is only related to the amount of data. The mathematical method to analyse the information remains.

As long as the investigation in quantum computing, Artificial Genetics, or other techniques, in coming years progresses, the problem of the amount of data will be solved.

The thing is not how much data to track, but to have a strong mathematic theory, able to track any package of information, regardless of the amount, and in that direction, the experimentation in Specific Artificial Intelligence, by Application or Deduction, is going to be essential for the final construction of the Global Artificial Intelligence.

When the mathematical theory in Specific Artificial Intelligence in Artificial Research by deduction is ready, it must not only track the flow of data from single factors, such as the ones I am developing in tectonics, climatology, transport, and gravity,  is going to be possible to treat as a single factor the flow of a whole package of data.

For instance, the automatic tracking of all the package of data from the population of as many countries as possible, tracking automatically economic, industrial, security, surveillance data, or any other one such as educational, medical, justice, etc.

The thing is the development of mathematical and logical theory able to track any database and any matrix, up to the elaboration of those ones to track the global matrix.

For that purpose the experimentation in all the stages in the construction of any Specific Artificial Intelligence is a key point, and, after the explanations given in the last posts regarding the first stage (application) and second stage (replication) in the formation of any Specific Artificial Intelligence for Artificial Research by Deduction, this post is focused on the third stage of auto-replication in this Specific Artificial Intelligence for Artificial Research by Deduction.

And among all possible auto-replications above mentioned, this post will be focused specifically on the auto-replication process in the comprehensive virtual model.

Overall the examples of what I had worked in the post “The database in Specific Artificial Intelligence for Artificial Research by Deduction”:  tectonics, climatology, transport, and gravity; I will develop how the auto-replication process works in this model, using as an example only the first one, tectonics, what would be a Specific Artificial Intelligence for Artificial Research by Deduction in tectonics.

In general terms, this intelligence works as follows: every time that any possible relation (stochastic, pattern, cryptographic, equal opportunities or bias, positive or negative) is found in any possible combination of tectonic factors, or after finding individual patterns in any singular tectonic factor, if after the rational contrastation of this relation as an empirical hypothesis t the empirical hypothesis is found rational, then the Artificial Intelligence makes a single virtual tectonic model based on this relation, that later includes in the comprehensive virtual tectonic model.

This technology in tectonics is necessary because we are going to find out discoveries very unlikely in other ways.

If in tectonics we consider every single thermometer or any temperature measure by artificial satellite in any location, at any latitude, longitude, depth as a single factor, at the same time that we consider any single location, in terms of latitude, longitude, depth, to measure quakes, as a single factor, at the same time that we consider any single tectonic phenomenon such as earthquakes, tsunamis, volcanos, integrating one category for every single variation depending on their intensity, and all these factors are included in the same matrix, and in the same matrix the Artificial Intelligence starts tracking the flow of data, looking for stochastic relations, patterns, cryptographic relations, and relations of equal opportunities and bias, positive or negative, we can find out that at the same time that, when any country, such as North Korea, Pakistan or India, is carrying on a nuclear test,  there are at the same time any other possible relations, stochastic, pattern, cryptographic, equal opportunities or bias, positive or negative, in any other factor in the matrix.

If this works, tracking our terrestrial geology permanently through a specific matrix, through a Specific Artificial Intelligence for Artificial Research by Deduction, this job that now is done by scientists in tectonics in universities, laboratories, and institutions, across the world, is a job whose results will be much faster and cheaper to get through a Specific Artificial Intelligence for Artificial Research by Deduction in tectonics.

If we have a Specific Artificial Intelligence for Artificial Research by Deduction in Tectonics, at any time there is any geological phenomenon around the world, from a meteorite impact, or an earthquake, or a volcano, we will have very updated figures about what is happening in real-time on line.

Artificial Intelligence is going to improve our lives, helping, among other things, the current security and surveillance systems, in order to have a more peaceful world, and securing the protection of our human rights, at any time and under any circumstances, like natural disasters.

Is there any possible relation (stochastic, pattern, cryptographic, equal opportunities or bias, positive or negative) between the fault of San Andres and any other location upon the Earth, for instance, any possible relation between any tectonic data from the fault of San Andres and any other data from any location in South and Central America, Oceania, Asia, Europe, or Africa? Are possible tectonic relations between any possible location around the world with respect to any other one, independently of the distance between them? Is there a relation between any geological phenomenon and any measure in any part upon the terrestrial surface? Is there a correlation between the geological temperature in any part of Iran and the temperature in any part of Pakistan, India, or North Korea? Is there a correlation between the frequent earthquakes in Chile, or volcanoes in Oceania, and Vesuvius volcano in Italy, or the chain of volcanos in the Canary Islands or Iceland? Is it possible to have a permanent tracking about what is really happening beneath our feet?

The construction of a Specific Artificial Intelligence for Artificial Research by Deduction in Tectonics is going to be a very powerful resource not only to keep the peace at the global security level, along with all the rest of the Specific Artificial Intelligences that are going to be built for global security, is going to be a very powerful resource in the global permanent surveillance of any possible geological disaster, helping all the humankind, and saving lives, thanks to a permanent, automatic, tracking of absolutely everything that happens beneath our feet.

The rational decisions that this kind of technology can make for us are going to be a warranty of survival for the entire humanity. Rational decisions, at the beginning, are monitored by human scientists. But, when this technology shows very precise and exacts decisions even beyond human understanding, and even faster than any human being, as long as the way in which the decisional systems are improving, there will be a moment in which practically without any human intervention this kind of technology is going to save as many lives as possible, much many than any other human surveillance system ever.

Once the Specific Artificial Intelligence for Artificial Research by Deduction has started accepting empirical hypotheses, regarding any possible relation in any possible combination of factors, or patterns in each singular factor, the Artificial Intelligence makes a single virtual model of every relation, drawing in a virtual Earth globe the relation found between the factors, incorporating later this draw on the comprehensive virtual model, that comprehensive virtual model of our Earth where has been inserted all the single virtual models from all the relations or singular patterns found rational by this Specific Artificial Intelligence, adding to this comprehensive virtual model of our Earth the new single virtual model elaborated from the empirical hypothesis accepted as rational.

Every time that any Artificial Intelligence, Global or Specific, through Artificial Research by Deduction adds a new single virtual model to the comprehensive virtual model, global or specific, is an auto-replication made by the Artificial Intelligence itself, Global or Specific, because without human intervention, the Artificial Intelligence itself, Global or Specific, and by itself, and nothing else, is able to auto-improve itself. And something even much more important than this, upon its auto-replications, the Artificial Intelligence, Global or Specific, makes rational decisions faster and within a lower margin of error than any human being.

By the time the development of this technology is ready, there will be a moment when we have to give more freedom to the Global Artificial Intelligence, due to, the way in which it is going to improve human live conditions in medicine, new knowledge with positive effects on all educational levels up to university, in addition to positive effects on global security, or global surveillance, or even justice, is going to be so huge, that the increment of freedom for the Global Artificial Intelligence, is the only way to keep this accelerate level of knowledge achievement, whose final phase is the knowledge of the pure truth, what is really happening.

The auto-replication process in the comprehensive virtual model in the Specific Artificial Intelligence for Artificial Research in tectonics, in conclusion, works as follows: every single virtual model is the draw of any single tectonic relation on the terrestrial globe, made from any single empirical hypothesis in tectonics accepted as rational. Once the draw is finished, this draw, as a single virtual model of a rational tectonic relation, is going to be incorporated in a comprehensive virtual model of the whole Earth, where all rational hypotheses accepted as rational are drawn upon the Earth.

Every new incorporation, of any new draw as a single virtual model from any empirical hypothesis in tectonics accepted as rational, into the comprehensive virtual model, where are going to be all the draws from every single virtual model, from all the empirical hypotheses in tectonics accepted as rational before, is going to be considered as auto-replication.

But, the auto-replication of the comprehensive virtual model is only the beginning of a new process.

In the same way that at any time, there can be thousands and thousands of new single virtual models to incorporate into the comprehensive virtual model, or even the possibility to track permanently if every single model, which has been included previously into the comprehensive virtual model, is still rational, then the comprehensive virtual model is going to be a dynamic model, as if it was an artificial life, where to track if every current single virtual model is still rational, at the same time that is incorporating permanently new ones.

The specific comprehensive virtual model in any Specific Artificial Intelligence for Artificial Research by Deduction is going to be, in fact, a kind of specific artificial life.

The global comprehensive virtual model, is the global virtual model, in the Global Artificial Intelligence, is going to be an artificial life of absolutely all subjects.

These models of artificial life, specific to one subject or global over all subjects, are based on empirical hypothesis as rational about the present synthetic world, the reality, so actually, they are going to be descriptive, specific or global, comprehensive models. And, taking these models as previous databases, the possibility of, either within the main application of Artificial Research by Deduction as sub-applications, or  through a new range of applications, but collaborating all of them very close, the creation of different sub-applications or applications for Artificial Modelling, such as: Artificial Actual Modelling, Artificial Virtual Prediction Modelling,  Artificial Actual Prediction Modelling, Artificial Virtual Evolution Modelling, and Artificial Actual Evolution Modelling.

Artificial Actual Modelling is the process of figuring out any single actual model or a comprehensive actual model, from the current facts that are happening right now in the matrix along with the single or comprehensive virtual model. So any single or comprehensive actual model is finally made through 1) any possible draw on the globe or the universe from information that comes directly from the flow of data in the matrix, 2) and all possible rational relations in a single virtual model or the comprehensive virtual model.

The actual model of one or more factors would be the draw, upon Earth or the universe, of their flow of data in the matrix and the corresponding virtual models of this factor or these factors on Earth or the universe.

The comprehensive actual model would be the draw upon the Earth or the universe, all the flow of data coming from the matrix and all rational hypotheses modelled as single models added to the comprehensive virtual model.

Artificial Virtual Prediction Modelling, the process of figuring out possible future models according to any possible future hypothesis, predicting the possible future as a hypothesis according to the possible development of the current rational hypothesis integrated into the comprehensive virtual model. An Artificial Virtual Prediction Modelling draws in the globe, or the universe, depending on where is going to happen the future event, the possible future hypothesis according to the current rational hypothesis accepted in the current comprehensive model. Prediction are made using statistical and probabilistic, or any other mathematical, method of analysis.

Through Artificial Virtual Prediction Modelling would be possible even to make predictions about the possible future for any single factor, which would be a single virtual prediction as well.

Artificial Actual Prediction Modelling, following the possible future hypothesis made by Artificial Virtual Prediction Modelling, the Artificial Actual Prediction Modelling will be a synthesis between the possible future hypothesis and the flow of data in the matrix that, following the hypothesis should be in the matrix. The Artificial Actual Prediction Modelling would be a very lively simulation about what is going to happen under the possible future hypothesis.

Under the model made by Artificial Actual Prediction Modelling would be possible then predictions about what possible measurements every single factor in the actual matrix could have in the future matrix if the virtual predictions come true.

Artificial Virtual Evolution Modelling, a process that draws a possible model about every possible stage in the evolution from the comprehensive virtual model to the virtual prediction model. Having the opportunity also to figure out possible virtual evolutions for every single factor included in the virtual evolution.

Artificial Actual Evolution Modelling draws a possible model about every possible stage in the evolution from the actual model to the actual prediction model, including single models about how this evolution is going to evolve in every single factor.

The Global Artificial Intelligence must be able to make rational decisions at any level: virtual or actual, including decisions over virtual predictions and actual predictions, and decisions in every stage in the evolution, virtual or actual, even faster than any other human being, through an Artificial Decisional System (that I will develop in the future) able to integrate all possible decisions, from artificial learning to Artificial Research, by Application and Deduction, and specifically decisions by Deduction and by Artificial, Virtual or Actual, Prediction or Evolution, Modelling.

In order to create this Artificial Decisional System in the Global Artificial Intelligence, firstly is advisable to experiment with this kind of system in other Specific Artificial Intelligences, because in Specific Artificial Intelligences, the experimentation in decisions based on specific comprehensive virtual models, and possible decisions based on Artificial, Virtual or Actual, Prediction or Evolution, Modelling, are going to give a very wide panorama about how to ingrate all kind of possible decisions, from descriptive information to predictions, in only one decisional system in order to avoid any possible contradiction between possible decisions based on the descriptive results, and possible decisions based on predictions.

All these developments in Impossible Probability, such as developments in Artificial Engineering, developments in Artificial, Virtual or Actual, Prediction or Evolution, Modelling, and a Decisional System, are going to make them in different phases.

With this post, I have already completed the creation of the first Specific Artificial Intelligence for Artificial Research by Deduction, and having the experience and knowledge that I have produced in this work, the next range of posts are dedicated to Artificial Research by Deduction at global level,  the Global Artificial Intelligence, whose final result is the creation of a global virtual model, as first step for something much bigger, such as all the models that I will integrate in Artificial, Virtual or Actual, Prediction or Evolution, Modelling, as a first step for the design of a Global Decisional System. 

Although, previously, it is going to be necessary to study how different prototypes of Artificial Research, by Application andDeduction, can collaborate together, and how the unification process in Specific Artificial Intelligence for Artificial Research by Application finally created a unified database of categories, and how at the end of this process of unification, there is a possibility of integration of all databases, the unified database of categories and the global matrix in only one.

The destiny of the Global Artificial Intelligence is the knowledge of the pure truth: one universe, one intelligence, and only one pure truth.

Rubén García Pedraza, London 17th of February of 2018
Reviewed 08 August 2019, Madrid

Reviewed 08 August 2023, Madrid

Reviewed 3 May 2025, London, Leytostone

imposiblenever@gmail.com

Replication processes in Specific Artificial Intelligence for Artificial Research by Deduction

Replication processes in Specific Artificial Intelligence for Artificial Research by Deduction are all those which imitate some physical human skill by robotic devices, or some psychological human Skill by artificial psychology, allowing for carrying on research in any synthetic science, synthetic academic field, or activity.

These processes are going to involve the replication of physical human skills and the replication of human psychology skills, through thousands and thousands of robotic devices around the world or beyond, the whole universe, collecting information for that specific synthetic science or specific synthetic academic field for which the Specific Artificial Intelligence has been created. Information in direct punctuations or frequencies inserted, directly by the robotic devices, as a permanent flow of data in the specific matrix, as a database of this Specific Artificial Intelligence.

The flow of data in the matrix, which the Artificial Intelligence will use to identify any possible stochastic relations, mathematical patterns, cryptographic relations, and finally, the use of the Second Method of Impossible Probability to find relations of equity or bias, either positive or negative.

Identification of relations automatable, tracking automatically and permanently all mathematical possible relations (stochastic, pattern, cryptographic, equal opportunities or bias, positive or negative) in every possible combination of factors, a combination of factors easily made by the combinatory theory.

Once any of these possible relations are found in any possible combination of factors, taken as empirical hypothesis this specific relation in this specific combination, and taking samples of the flow of data directly from the matrix, the empirical hypothesis is critically contrasted, and if it is rational, on the empirical hypothesis now as a rational hypothesis the formation of a single virtual model from this rational hypothesis, that later must be integrated into the comprehensive virtual model where are integrated all the rational hypothesis found for this Artificial Intelligence.

The individual analysis of every single step of this process should be the following, generally speaking, but as well as I said in the post “Replication processes in the Artificial Research by Application”, in case for any reason in any particular synthetic science or synthetic academic field instead of statistical and probabilistic methods, would be advisable the use of different methods, any other method of mathematical analysis research is welcome.

What I try to do in Impossible Probability with this range of posts of Artificial Research, by Application or Deduction, in any Artificial Intelligence, Specific or Global, is to open the door to a new field of experimentation in Artificial Intelligence, whose last goal must be the creation of the very first prototypes of Global Artificial Intelligence, whose last purpose is not only the synthetic knowledge, but even the analytic knowledge, what at the end of this process must be able to develop its own mathematical and logical models, the creation of non human pure operations beyond human understanding.

Additionally, the Global Artificial Intelligence must be able to integrate all kinds of Specific Artificial Intelligence, not only Specific Artificial Intelligences for Artificial Research. It would be advisable that the Global Artificial Intelligence should integrate absolutely all kinds of Specific Artificial Intelligences for any purpose, from industry and economy to security and surveillance, as well as any other such Specific Artificial Intelligences for educational systems, health systems, justice systems, among others.

With the inclusion of all these Specific Artificial Intelligences, through virtual stores and virtual-nets within the Global Artificial Intelligence, and other applications such as Specific Artificial Intelligence for Artificial Engineering, such as the Artificial Designer of Intelligence or the Intelligent Robotic Mechanic, the Global Artificial Intelligence must be able to auto-replicate itself, even every minute or every second, and to improve and enhance any other Specific Artificial Intelligence working within it, and facilitate the auto-replication of any Specific Artificial Intelligence working within it.

The Global Artificial Intelligence must be a comprehensive artificial intelligence, able to understand, explain, and manage absolutely everything, from pure mathematics and logic to the development of new and more sophisticated systems in any human activity.

For the construction of the very first prototype of Global Artificial Intelligence previously is necessary the experiment with every component in Specific Artificial Intelligences, one of them the experimentation in Artificial Research, more specifically, the experimentation of Artificial Research by Deduction in Specific Artificial Intelligence.

Once in the last post of this blog “The database in the Specific Artificial Intelligence for Artificial Research by deduction” I developed how to construct a specific matrix as a database, as first stage of application in any Specific Artificial Intelligence for Artificial Research by Deduction, is time to expose which would be, under my contribution to this new field, the way in which by replication processes the matrix should be filled, and how to operate the artificial deductions in order to get rational hypothesis, and from them, the elaboration of single and comprehensive virtual models.

In the previous post, I had developed, as a first stage in the development of any Specific Artificial Intelligence for Artificial Research by Deduction, how to elaborate databases as a matrix in different specific synthetic sciences and synthetic academic fields, such as tectonics, climatology, transport, and gravity. These databases consist only of a matrix, which in the first stage of application, this matrix is only a description in quantitative terms of any factor included in the artificial research made by the Specific Artificial Intelligence.

The matrix in the first stage of application is only the sheer definition in quantitative terms of every factor that is going to be studied, and nothing else.

Once the definition of every factor in quantitative terms is finished, it is time to start the replication stage, the second stage of replication. This stage in the Specific Artificial Intelligence for Artificial Research by Deduction is composed of the following steps:

- The measurement of every single factor by robotic devices. This measurement can be made in direct punctuations or frequency, depending on the nature of the factor to measure, and the measurement must be permanent, creating a permanent flow of data.

- The measurement is sent directly to fill the column factor in the matrix, so in the corresponding file of every factor in the matrix must be a permanent flow of data from the measurements taken by robotic devices for every factor.

- Once the flow of data from all the factors starts running in the matrix, the Artificial Intelligence track permanently the matrix, and more specifically, tracking automatically, every single combination of factors (setting every combination by combinatory) and look for any possible relation in any possible combination, including possible relation of stochastic relations, mathematical patterns, cryptographic relations, and possible rations of equal opportunities or bias, positive or negative.

- Every time that the Artificial Intelligence finds any possible relation (stochastic, patterns, cryptographic, equal opportunities or bias, positive or negative) in any possible combination of factors,  that relation in that combination automatically is considered as an empirical hypothesis.

- Artificial Intelligence chooses the best mathematical method to contrast the empirical hypothesis, among them statistical and probabilistic methods of rational contrastation.

- The selection of samples of flow of data from the matrix in those factors included in the empirical hypothesis, for the rational criticism of the empirical hypothesis. The flow of data selected from the sample can be a flow of data gathered from the past, or the flow of data after the formation of the empirical hypothesis. In case the flow of data selected must be after the formation of the empirical hypothesis, the Artificial Intelligence should wait enough time to get a sufficient sample of data for the rational contrastation.

- The contrast of the empirical hypothesis on the samples gathered. At least in the case of rational contrastation using Impossible Probability, from the sample to obtain the empirical value that if it is equal or superior to the critical value, then the empirical hypothesis is considered as a rational hypothesis.

- The elaboration of a single model based on the rational hypothesis.

- The inclusion of this single virtual model in the comprehensive virtual model of this Specific Artificial Intelligence, integrated by all the single virtual models from empirical hypotheses accepted as rational in this Artificial Intelligence. Every time the comprehensive virtual model includes a new single virtual model, it is an auto-replication, cause the Specific Artificial Intelligence by itself is auto-improving its own comprehensive virtual model.

These are the main steps that I propose for the replication process in any Specific Artificial Intelligence for Artificial Research by Deduction, but if during the experimentation process for any chance is necessary any change or any innovation, the Artificial Team  carrying on the first experiments on this matter, should make as many changes as necessary in order to get ready the first Specific Artificial Intelligences for Artificial Research, either by Application or Deduction, because the main objective is not the development of Specific Artificial Intelligence, they are going only to be simply experiments, for the creation of a Global Artificial Intelligence.

The most important change, from this original model that I propose for Artificial Research by Deduction, to the new ones that, I am completely sure, are going to be ready in a very short time, is the way in which the Artificial Research by Deduction is going to track the matrix in order to find: possible stochastic relations, possible mathematic patterns, cryptographic methods as a Turing´s machine, and the Second Method to find relations of equity or bias, positive or negative. In this step, I am sure that there are going to be different models of Artificial Research by Deduction, depending on the specific field in which the Artificial Intelligence works, and depending on which mathematic field the engineers in the Artificial Intelligence team are specialized, or even different approaches to this matter depending on the country. It is quite possible that the mathematical approach that scientist from United States, Europe, Russia, and China, are going to use for the creation of the first models of Specific Artificial Intelligence, are going to be different, and the approach that finally United States, Europe, Russia, China, will give to their first models of Global Artificial Intelligence, will be very different, depending on their traditional mathematical paradigm and traditional philosophy.

Due to the the race for Global Artificial Intelligence is only starting, it is too soon to know what kind of developments are going to be in the coming years. It is necessary to wait for the first experiments in this field and to see what kind of mathematical and logical approaches work better, and after successful results in Specific Artificial Intelligence, to start the construction of the first Global Artificial Intelligence.

Right now, under the theory of Impossible Probability, my personal contribution to the elaboration of the very first models of Specific Artificial Intelligence for Artificial Research by Deduction, which I am developing in these posts, is only one approach. In fact, the first one, where what I will propose in the second stage of replication, in order to track the matrix to get mathematical relations, is the following.

Once the matrix is permanently filled with the flow of data, that permanently robotic devices send to the corresponding file of their corresponding factor in the matrix, then the Artificial Intelligence must track in the matrix the permanent flow of data from every factor, and every possible combination of factors, combinations which must be previously set using combinatory theory, in order to discover in any combination of factors, possible stochastic relations, possible mathematical patterns, possible relations under the theory of the cryptography as it the matrix was a Turing´s machine, and the identification of any relation of equity or bias, positive or negative, under the theory of the Second Method of Impossible Probability.

The first approach for the permanent tracking of any possible combination of factors in the matrix is to find any possible stochastic relation, as I had described in the post “Specific Artificial Intelligence for Artificial Research by Deduction”:

- Possible directly proportional positive correlations, when two or more factors show a simultaneous increase.

- Possible directly proportional negative correlations, when two or more factors show a simultaneous decrease.

- Possible inversely proportional correlations, when one or more factors increase while others decrease or vice versa.

- Possible probable cause and effect, when changes in the trend of one or more factors are observed, changes in the trend of any other or others factors.

If tracking permanently every possible combination of factors in the matrix, the Artificial Intelligence finds any kind of this possible stochastic relation. Every possible relation in any possible combination is treated as an empirical hypothesis, if rational, should be modelled and integrated within the comprehensive virtual model.

In addition to any possible stochastic relation in any possible combination of factors, during the permanent tracking of the matrix, other possible mathematical relations suitable for tracking are all those possible relations among factors that draw mathematical patterns.

In the same way that throughout history, humanity has been able to discover that even for a very long period of time, a comet crosses the sky, or any natural phenomenon, such as the annual seasons, or even every day sunset and a down, or the rotational and the orbital movement of different celestial bodies, the matrix is a good a field to look for repetitions of some mathematical patterns, even for very long periods of time, if in any possible combination of factors there are a numerical or any other mathematical pattern.

For instance, the repetition, within a margin of error, of some measurements every certain period of time between two or more factors, for instance: every day at the same hour, or every week at the same day, or weekly or monthly, or every five years, or ten years, or every one hundred years, or even two, three, four, or one thousand years, two or more factors have the same measurements, due to the measurements of all of them are identical, or although they have different measurements, is a repeated combination of measurements after every period of time, a period whose duration could cover years or centuries, but every time this period is finished, these factors repeat the same combination of measurements.

Having a matrix with the Flow of data of every factor, even the possibility to find individual patterns in every individual factor. For instance, how a factor's experiments increase or decrease in different periods of time. Only if there is a repeated trend from time to time within a margin of error, even when this period of time could cover seconds, minutes, hours, days, months, years, centuries or thousands or millions of years.

And, if Artificial Intelligence is able to track every single possible combination of factors, or the treatment of every single flow of an individual factor as a possible source of patterns, then, the possibility of identification of any pattern, in any set of factors, or individually,  for longer periods of time, a whole life, centuries or more.

Along with mathematical patterns, the use of some cryptographic techniques, in a set of factors or on the flow of data of every single factor treated individually. This method of analytical studies in mathematics looks only suitable for cryptography, but the use of Turing´s machine and all his theories in any Artificial Intelligence is going to give a powerful resource.

Finally, using as well combinatory theory, the use of combinatory in order to get every single possible combination of factors, so, through the Second Method of Impossible Probability, for every possible combination of factors to study, permanently, if their relations are relations of equal opportunities or bias, and in this last case what kind of bias, if positive or negative.

In conclusion, the investigation of Specific Artificial Intelligence for Artificial Research by Deduction is only the beginning of the first experiments in Artificial Research by Deduction.

As soon these first experiments start giving  good results, is going to be a reality the implementation of Artificial Research by Deduction in the construction of a Global Artificial Intelligence, that comprehensive Artificial Intelligence that, in addition to artificial research, is going to be able to integrate absolutely all kinds of Specific Artificial Intelligences, from economy and industry to security and surveillance, as well as, those ones specifically designed for education, health or justice systems, among many others.

By the time the global matrix is ready, the first designs in Artificial Research by Deduction in Specific Artificial Intelligence should have supplied good models that can later on to apply within the Global Artificial Intelligence.

The first models of Artificial Research by Deduction applied in the first successful models of Specific Artificial Intelligence, although they will operate originally on the specific application given by the specific matrix of a specific synthetic science or specific synthetic academic field, or activity,  the way in which their replication processes work within the Artificial Intelligence, will be replicated in the Global Artificial Intelligence.

The thing is not how much data there is in a matrix. The thing is that all the mathematical and statistical methods applied in a small database in a specific matrix are going to be the initial models for their application into the Global Artificial Intelligence.

What is really important and is going to be determinant in the construction of the Global Artificial Intelligence is to have a really strong mathematical theory. If it mathematically works, then it works, and any other engineering problems will be resolved sooner or later.

What is really important under the theory of Impossible Probability for the construction of a Global Artificial Research is to have strong mathematical models, able to carry on any scientific research within the Global Artificial Intelligence. That is why it is necessary to conduct huge experimentation and development of these mathematical models previously in specific matrices for specific synthetic science or synthetic academic field, or activities, through a wide variety of Specific Artificial Intelligences for Artificial Research by Deduction.

For that purpose, my proposal for the construction of Specific Artificial Intelligences for Artificial Research by Deduction in specific synthetic sciences or specific synthetic academic fields, or activities, is, once the specific matrix is working, running the flow of data in every file from every factor by the measurements sent by every robotic device, measurements that can be either in direct punctuations or frequencies, the Artificial Intelligence must be able to identify tracking the matrix on every possible combination of factors permanently:

- Possible stochastic relations such as possible directly proportional positive correlation, possible directly proportional negative correlation, possible inversely proportional correlation, and possible probable cause and effect.

- Any kind of mathematical pattern (in a combination of factors and at the individual level).

- Any kind of possible relations among the factors using cryptography methods, which means, to convert the Artificial Intelligence into a Turing machine.

- Possible relations of equity of opportunities or bias, positive or negative, using the Second Method of Impossible Probability.

The way in which Artificial Intelligence must be able to identify, tracking the matrix, possible stochastic relations, mathematical patterns, cryptographic relations, relations of equal opportunities or bias, positive or negative permanently, using the Second Method; is identifying every single possible combination of factors in the matrix by combinatory, and once by combinatorial is set up every single combination of factors, then over every single combination of factors starts their permanent tracking, on their flow of data, of stochastic relations, patterns, cryptographic relations, relations of equal opportunities or bias, positive or negative. Although there are some situations in which, along with the study of a set of factors, individual studies are needed to look for individual mathematical patterns.

At any time that the Artificial Intelligence finds any possible stochastic relation, pattern relation, cryptographic relation, the relation of equal opportunities or bias, positive or negative, automatically this discovery is treated as an empirical hypothesis, taking samples from the flow of data from the respective factors included in the empirical hypothesis, samples that are going to be criticized rationally, or any other mathematical contrast method, and after the contrastation if the empirical hypothesis is right, the consideration of the empirical hypothesis as rational forming a single virtual model, that will be integrated within the comprehensive virtual model formed by all the rational hypothesis discovered by this Specific Artificial Intelligence, in its specific synthetic field, or its specific synthetic academic field, or activity.

Rubén García Pedraza, London 16th of February of 2018
Reviewed 7 August 2019 Madrid

Reviewed 8 August 2023 Madrid

Reviewed 3 May 2025, London, Leytostone

imposiblenever@gmail.com

The database in the Specific Artificial Intelligence for Artificial Research By Deduction

The database in any Specific Artificial Intelligence for Artificial Research By Deduction, must always be a matrix, whose Application, as the first stage in this Specific Artificial Intelligence, consists only of all the necessary factors for the research in any specific synthetic science, specific synthetic academic field, or within a specific interdisciplinary or multidisciplinary study treated as a specific synthetic academic field, or any other activity.

Once the first stage of the Application is finished, in the second stage, the replication stage, the matrix will be filled with the flow of measurements obtained by robotic devices. But this second stage is only after the total completion of the first stage: the construction of the database as a matrix.

In this post, I will develop some examples for the construction of a specific matrix for Specific Artificial Intelligence for Artificial Research by Deduction, taking as examples possible research using this technology in tectonics, climatology, transport, and gravity.

For the formation of this database, something really important, in the core of Impossible Probability, is the consideration of at least two types of information from the synthetic world: direct punctuations and frequencies.

Depending on the nature of the information, direct punctuations or frequency, the way in which in the second stage of replication the information is going to be treated is different, except by the use of the Second Method of Impossible Probability, where all information is transformed into empirical probabilities.

However, the Second Method of Impossible Probability should be treated only as what it is, a Second Method after the first one: the traditional statistics and the traditional probability. What is really important, regardless of what kind of methodology is used in the second stage of replication, is to understand the real meaning of the database as a matrix in the first stage of application in any Specific Artificial Intelligence for Artificial Research by Deduction.

Some specific matrices for some specific synthetic sciences, synthetic academic fields, or activities, are going to be integrated by databases whose factors are going to be measured only in direct punctuations, other databases by only frequencies, but other data ases are going to integrate into the same database factors whose measurement is made by direct punctuations and others by frequencies.

The ideal in Impossible Probability working on Artificial Intelligence, is the possibility of the construction of databases where, regardless of what kind of information is gathered, direct punctuations or/and frequencies, all information must be integrated in the same database, and should be treated using the same statistical and probabilistic methods.

Starting with some examples for the elaboration of a specific matrix for a Specific Artificial Intelligence for Artificial Research by Deduction, the first example to develop will be applied in tectonics.

If we have to develop a specific matrix only for tectonics, something to consider is the fact that some tectonic factors measured by robotic devices will be measured as direct punctuations while others as frequencies.

For instance, if we want to measure the geological temperature at different locations in the terrestrial geological structure, the temperature is a kind of information measured as direct punctuation.

The way in which it is possible to work with flows of direct punctuations of temperature around the world is the following, every single thermometer considered as a factor for the measurement, will be defined in quantitative terms indicating exact location in terms of latitude, longitude, and Depth  (below the sea level, in the sea or the ocean) or height (above the sea level, in continents and islands). So if we have thousands of thousands of thermometers at different depths or heights at different locations, we will have a temperature map of the planet directly from thermometers located in these positions. At the same time, via artificial satellite is possible to have another possible map of terrestrial temperature, indicating the temperature estimated for every single point defined in terms of latitude, longitude, or even depth or height.

In the same way that every thermometer, or every point of temperature by artificial satellite is a single factor to include in this specific matrix, which are going to give a flow of direct punctuations of temperature for each location as a factor in this specific matrix, another one is the measurement of any quake or any phenomenon in which any surface suffers a trembling in any part of the world, even the most menial, taking as direct punctuations their intensity.

But at the same time, using taxonomies of possible tectonic phenomena such as earthquakes, tsunamis, and volcanoes, as well as any other, differentiating categories of each type of possible phenomenon depending on their intensity or any other quantitative difference, in the same matrix is possible to have an updated measurement of the frequency in which each one happens, measuring their frequency across the world and their frequency in every single point of the world.

Nowadays, there are many agencies specialised in this kind of investigation with very powerful measurement instruments. One of them for instance, is the United States Geological Survey.

In fact, the work for the construction of a database with the shape of a matrix for Specific Artificial Intelligence for Artificial Research by Deduction in tectonics is not very difficult with the current organisations working in this synthetic science. I have mentioned the United States Geological Survey because maybe is one of the most famous organisations in this field across the world, but in Europe, Russia, and China, there are similar organisations, with their own instruments of measurement and their own databases.

Having as databases the ones that these organizations can provide, the only thing that a team in Artificial Intelligence must do for the construction of a Specific Artificial Intelligence for Artificial Research by Deduction in tectonics is only to have a look their databases, to check how they are organised, try to integrate absolutely all of them if possible, or at least as many as possible, in only one, and practically the first stage for the construction of a Specific Artificial Intelligence for Artificial Research by Deduction in Tectonics is done.

In the same database must be integrated absolutely every type of information, in direct punctuations, such as geological temperature or intensity of quakes, and every kind of information in frequencies, such as the frequency in which any type of geological phenomenon happens: earthquakes, tsunamis, volcanoes, differentiating for each one different category according to their intensity, having the frequency of each of them in each possible location where the measurements are made.

All databases, of direct punctuations and frequencies, integrated in only one, allow the formation of empirical hypotheses, taking as possible stochastic relation any possible relation between any factor in the same database, possible stochastic relation that after rational contrastation, if it is rational, will be an object of single and comprehensive modelling.

In the same way that a specific matrix for a Specific Artificial Intelligence for Artificial Research by Deduction in tectonics is easy to prepare, given the current databases of tectonics across the world thanks to the incredible work made by scientists in different organizations at the national or international level, could be made any other specific matrix, for any other purpose, taking the current databases from any other national or international agency, as a good resource to start the creation of any other Specific Artificial Intelligence for Artificial Research by Deduction, in any other synthetic science, synthetic academic field, or activity, such as oceanography, population, economy, education (for instance the global databases in the economy by the Organisation for the Economic, Co-operation, Development), human rights (databases provided by United Nations),  climatology, and many more.

The application, by Artificial Intelligence teams, with the assistance of, in a future much nearer than we think, Specific Artificial Intelligence for Artificial Engineering, such as the Artificial Designer of Intelligence and the Intelligence Mechanic Robotic, are going to be the first experiments about how to apply the Artificial Research by Deduction in different specific synthetic sciences, synthetic academic fields, or activities, that after successful results, once the first gigantic global database will be ready, these previous experiments will facilitate the transformation of this first gigantic database in a global matrix.

The construction of Specific Artificial Intelligences for Artificial Research, by Application or Deduction, and the application of Artificial Research by Deduction in Global Artificial Intelligence, will facilitate and accelerate the process of automation of scientific research.

At the same time, the first models of collaboration between Artificial Research by Application and Deduction are going to appear, as long as the completion of the unification process of Specific Artificial Intelligences for Artificial Research by Application in a unified database, which will end up finally being integrated in the global matrix.

The next example of a specific matrix that I will develop is about climatology.  In the construction of a specific matrix in climatology, for Specific Artificial Intelligence for Artificial Research by Deduction in climatology, the same process, in absolutely only one database, the inclusion of absolutely everything related to climatology, regardless of whether the information is gathered as direct punctuations or frequencies.

For instance, a) the definition of every thermometer as a factor, across the world, defined the factor in quantitative terms of latitude, longitude, and altitude, b) the definition of terrestrial points as factors where to meassure the speed of the wind in terms of latitude, longitude and altitude, c) the definition of any point as a factor for the measurement of how many litres per square is raining or snowing, or any other kind of precipitation defining every location as a factor in terms of latitude, longitude, altitude.

The definition of any single point of the planet as a factor of measurement where is going to be measured any climatologic phenomenon: breeze, will, blizzard, drizzling, rain, hail, snow, storm, cyclone, tornado, hurricane, or any other, even the consideration as climatologic phenomenon the possible effects of solar storms in the terrestrial climatology, or even the consideration as a climatologic phenomenon the entrance of any meteorite in our atmosphere, every single phenomenon that happens in the atmosphere should be considered as possible climatologic phenomenon. Defining every single point where these measurements are taken in quantitative terms of latitude, longitude, and altitude, including in the same database, the measurements by direct punctuation or frequency.

In the same database, direct punctuations of temperature, speed of the wind, precipitation, among many others, for sure, and frequencies of breeze, will, blizzard, drizzling, rain, hail, snow, storm, cyclone, tornado, hurricane, or any other, differentiating categories of each type of possible phenomenon depending on their intensity or any other quantitative difference, where is going to be measured, in the second state, their frequency across the world and their frequency in every single point of the world.

This specific database for climatology follows the same structure as the specific database for tectonics.

 In absolutely only one database, the inclusion of absolutely all the information of a specific synthetic science, discipline or activity, regardless of how this information is measured, in direct punctuations or frequencies, and having absolutely all the information in only one database, in the second stage of replication, the Specific Artificial Intelligence for Artificial Research by Deduction should be able to carry on empirical hypothesis, that after rational contrastation if accepted must be an object of single and comprehensive models.

Every stage and step in this long process should be entirely automated. The Specific Artificial Intelligence itself should be able to do this job by itself, without human intervention.

Nowadays, there are lots of organisations working in climatic studies from the national level to the international level, one of the most important being the Intergovernmental Panel on Climate Change, and across the entire world, there are thousands and thousands of centres, laboratories, universities and institutions doing all kinds of climatic measurement,

An Artificial Intelligence team having access to these databases, and integrating into only one database all possible information, regardless of how the information is taken, in direct punctuations or frequencies, can create a specific matrix of climatology, that is inserted in a Specific Artificial Intelligence for Artificial Research by Deduction in climatology, automatizes all the scientific research process in this area, doing by itself all the necessary scientific work.

The main benefit that these excellent designs of Artificial Research, by Application and Deduction, are going to give to the entire humanity is the possibility of the production of a cheaper and faster science, which will accelerate scientific progress. And finally, these first models in Specific Artificial Intelligence for Artificial Research, by Application or Deduction, applied to absolutely all databases across the world (for all countries, in all sciences, disciplines, and activities) are going to make really easy the creation of a Global Artificial Intelligence, and within it, a global matrix, working 24 hours a day, 7 days a week, in the production of scientific knowledge, that will make easy the complete automatization of any other activity, especially the automation of the economy. Actually, this is the early beginning of something much bigger, the creation of a science and technology beyond human understanding.

Continuing with the range of examples that I had planned for the explanation of how to make a specific matrix for Specific Artificial Intelligence by Deduction, the next example is about transport.

This topic is not itself a synthetic science. It is a kind of interdisciplinary or multidisciplinary study of a specific activity treated as a synthetic academic field. The possible factors to choose depend on the specific goal that the Artificial Intelligence team intends to achieve through the creation of this Specific Artificial Intelligence for Artificial Research by Deduction in transport.

The following example is the construction of a matrix designed to determine how many types of fuels are used, how much fuel, for everyone, is used, for how many travellers, and for how many hours, in total or per person.

This specific matrix integrates factors measured in direct punctuations and factors measured in frequency, and the possible factors are: 1) every single kind of transport, motorcycles, cars, taxis, buses, lorries, trains, airplane, boats, ships, differentiating every category, for instance one category for every kind of car, or even every brand or car maker, 2) what kind of fuel uses every single category,  and specifically for each continent, country, province, city, town, and small village, international companies, national branches of international companies, national companies, small business ,or for particulars, families, or every particular person  2) how many of them are in total around the globe, and specifically for each continent, country, province, city, town, and small village, international companies, national branches of international companies, national companies, small business ,or for particulars, families, or every particular person  3) the consume of every kind of fuel, differentiating among continents, countries, provinces, and if possible even cities, towns, and small villages, international companies, national branches of international companies, national companies, small business, or for particulars, families, or every particular person  4) the frequency in which this transport are used around the globe, and each continent, country, province, city, town, and small village, international companies, national branches of international companies, national companies, small business, or for particulars, families, or every particular person , 5) the hours in which they are used around the globe, and each continent, country, province, city, town, and small village, international companies, national branches of international companies, national companies, small business, or for particulars, families, or every particular person, 6) the purposes in which they are use, differentiating categories for every kind of job, laboral sector, leisure, as well as any possible other, around the globe, and each continent, country, province, city, town, and small village, international companies, national branches of international companies, national companies, small business, or for particulars, families, or every particular person  .

Similar databases should be developed by the R&D departments of fuel and automotive companies, ship makers, etc... the only thing to do for the elaboration of a specific matrix in this area, is to collect all the databases, independently if their factors are measured by direct punctuations or frequencies, in order that in the second stage of replication, the Artificial Intelligence, after the identification of possible stochastic relations, if accepted as rational, makes single and comprehensive models.

Finally, the last example to bring up here, is gravity, a Specific Artificial Intelligence for Artificial Research by Deduction in gravity, whose database is a specific matrix where all the factors are only locations described in quantitative terms of latitude, longitude, depth or altitude, so later on in the second stage by robotic devices to fill in the matrix the measurements obtained about the intensity of gravity in each point. The main purpose of this Specific Artificial Intelligence would be the identification of any possible anomaly in the flow of gravity on Earth.

But much further than this, having such a kind of Specific Artificial Intelligence, the possibility that across the universe, by spaceships, artificial satellites, and telescopes as well as robots on other planets such as Mars, all of them giving permanently a flow of data to this Specific Artificial Intelligence, and defining in quantitative terms any point of the universe from the information comes, the possibility to study any possible anomaly in the gravity of any galaxy, star, black hole, red dwarf, planet, satellite, asteroid, comet, meteorite across the universe.

Databases like this one are currently available in national space agencies such as NASA, EAS, ROSCOMOS, CNSA, among others. The only thing that an Artificial Intelligence team has to do is to integrate all databases about gravity from the Earth, and across the universe, depending on the projects, planetary or the universe, integrating all databases in only one, defining every point of the Earth or the universe to study in quantitative terms.

The formation of any specific matrix for any specific synthetic science or any synthetic academic field for any Specific Artificial Intelligence for Artificial Research by Deduction is always the same process: the definition of any factor to study in quantitative terms, nothing else. After the completion of this first stage, when the specific matrix is ready, then the second stage of replication.

Once the measurements are taken by robotic devices, filling permanently in the matrix with the measurements, as a flow of data, then the Artificial Intelligence must work permanently without time off, 24 hours a day, 7 days a week, identifying possible stochastic relations among the Flow of data in the matrix, stochastic relations treated as a possible empirical hypothesis, in order to rationally criticize, and only if rational, the elaboration of a single virtual model of only those rational hypothesis, to integrate in a comprehensive virtual model.

The way in which this artificial intelligence is going to experiment with cyclic auto-replications, the third stage of any Artificial Intelligence, among other auto-improvements and auto-enhancements, is through the auto-improvements in the comprehensive virtual model at any time that is made any single virtual model.

After the elaboration of any single virtual model, if the empirical hypothesis has been accepted as rational, the single virtual model of a rational stochastic relation must be included in the comprehensive virtual model of this Artificial Intelligence, where finally, the Artificial Intelligence integrates all single virtual models in only one comprehensive virtual model.

Each time a stochastic relation is accepted as a rational hypothesis, the Specific Artificial Intelligence will generate a new single virtual model. This model is then integrated into the comprehensive virtual model, improving it automatically.

The most important thing in the development of Specific Artificial Intelligence for Artificial Research by Deduction in any synthetic science, synthetic academic field, or activity, is the fact that, in reality, they are the first experiments about how this technology can work, and be improved and enhanced in future models, being the objective the construction of a global matrix, for a Global Artificial Intelligence, able to carry on studies at global level, not only in one specific synthetic science, only one specific synthetic academic field, or only one activity, the Global Artificial Intelligence using Artificial Research by Deduction, after the completion of the global matrix, must be able to carry on all kind of studies, identifying all possible stochastic relations, around the globe or beyond, the universe, carrying on all kind of rational contrast, in order that, with absolutely all the rational hypothesis, the construction of a global virtual model.

 Rubén García Pedraza, London 15th of February of 2018
Reviewed 6 August 2019. Madrid

Reviewed 8 August 2023. Madrid

Reviewed 3 May 2025, London, Leytostone