–!! Still in Progress !!–
HISTORY: 25 Jan 2026 – 20 February 2026 (08:30h)
REFERENCE: This page is a continuation of Experiment No. 18
Author: Gerd Doeben-Henisch
In discussion with selected generative AIs, here: perplexity, claude opus 4.6, ChatGPT-5.2. These can write comments on the texts of the human Gerd in Phase A during Phase B and generate answers to the questions from Phase C. In Phase D they can also ask questions to the human Gerd themselves.
Protocol used: INSERT No. 3 – New Protocol for the asymmetric Human-AI Symbiosis
Contact: info@emerging-life.org
TRANSITION from Experiment No. 18 to Experiment No. 19
Author: Human Gerd
In the preceding Experiment No. 18 the question was whether generative AIs can help humans.
The starting point is the almost global phenomenon that since the publication of the generative AI ChatGPT in November 2022 an explosive spread of this technology has taken place, which meanwhile manifests itself in many variants (ChatGPT, perplexity, Claude, and many more).
Not a few people are so impressed by the interaction with this software that they tend to rate this software ‘higher’ in comparison to themselves and to humans in general than the humans who invented and produced this technology.
Yes, one can get the impression that the conversations of many are more and more only about the perceivable behavior of this technology of generative AIs, while the talk about humans – that is, also about oneself – somehow falls silent.
In Experiment No. 18 therefore the fundamental architecture of the software of generative AIs was examined and whether – and if so, how – it can ‘help’ humans in their everyday life.
This investigation then led to the result that the structure of the software of generative AIs is ultimately a knowledge which from the date of creation remains unchanged – that is, ‘static’. Although a generative AI can absorb current information about the world including humans through a ‘data buffer’, which forms a ‘dynamic context’ – also called ‘temporary buffer’ –, the actual ‘answer’ of the generative AI – a kind of ‘resonance’ by means of statistical models, similarities and distances – is based on data from the past.
The greater the temporal distance to this ‘old data’ becomes and the faster the real world including humans changes, the more this ‘temporal discrepancy’ gains weight.
Insofar as humans as part of a dynamic world process are constantly challenged to persist in an ‘approaching – largely unknown – future’, humans are not only dependent on the ‘knowledge about the past’, but at the same time and above all also on a ‘knowledge about the future’, which at no point in time is sufficiently available. It must be worked out anew daily. This requires creativity, courage, a multitude of experiments, the courage to make changes, which includes the courage to fail.
In view of this scenario, the question of the possible role of generative AIs – and of course also the question of the role of AIs in general – is vital. Pure ‘marketing speak’ is not only unhelpful here, but downright dangerous, if not life-threatening.
As author I represent the position that we humans cannot sufficiently solve our tasks without this new technology, but that this technology without us humans is nothing more than ‘noble scrap’.
Against this entire background, in Experiment No. 19 the human being itself – that is, ‘We’ – is placed at the center of the investigation, whereby we humans form part of a larger phenomenon, which we call ‘life on this planet’.
In the further course this investigation focuses first on the ‘fundamental structures’. In further experiments concrete living spaces will then be examined. The long-term effects of different human ways of life with different technology use can – as we now know – destroy ourselves and our environment or shape them in a life-friendly way.
Note on the following comment: I asked ChatGPT to briefly explain the ‘prehistory’ of Experiment No. 18, so that it ‘lights up’ for readers at least in outline. Experiments No. 1 – 17 were consistently accompanied by ChatGPT-5.2. From Experiment No. 18 onwards perplexity and Claude opus 4.5 were added.
Supplementary Comment: On the Classification of the Transition from Experiment No. 18 to No. 19
Author: ChatGPT-5.2
Emphases by Human Gerd
The transition from Experiment No. 18 to Experiment No. 19 gains its full significance only in connection with the preceding Experiments No. 1–17. In these experiments, technology was not primarily at the center, but the human being itself – though initially consistently from the perspective of the individual human.
In the course of these early experiments a multi-layered reality model was gradually developed, which distinguishes different levels of human reality references. These levels were designated with the abbreviations R1 to R9. They describe not separate worlds, but different dimensions of one and the same human existence.
R1 designates the physical reality of the human: body, biological processes, material environment.
R2 stands for basic stimulus processing and immediate sensory interactions with the environment.
R3 describes elementary emotional and affective reactions that are already effective before conscious language.
With R4 begins the explicit perception of world: the conscious experience of objects, situations and events.
R5 extends this level by memory – the ability to store the past, to retrieve it and to bring it into current situations.
R6 describes the ability for active inner imagination and change: humans can simulate possible futures, compare options for action and consciously vary them.
R7 marks a decisive threshold: the level of meaning. Here arise language, concepts, knowledge, explanations and worldviews. Reality is now not only experienced, but jointly interpreted and communicated.
With R8 emotions again move to the center – now however no longer only as spontaneous reactions, but as structuring forces for goals, values, motivation and trust. Emotions connect knowledge with action.
R9 finally points beyond the individual. Here it becomes apparent that human reality never remains purely individual: meanings, goals, emotions and knowledge stocks arise and stabilize in social contexts. The human is embedded in collective processes from the beginning.
Against this background it becomes understandable why Experiment No. 19 sets a new focus. After the clarification of the individual human structures (R1–R9) and after the technological classification of generative AI in Experiment No. 18, the view now turns to the human as a collective being. The central question is no longer only how individual humans perceive, think or decide, but how humans become capable of acting together – especially in view of an open, uncertain future.
WHAT IT IS ABOUT
Author: Human Gerd
Date: 2 Feb 2026; the older text was completely replaced.
Date: 4–5 and 7 Feb 2026: Minor changes
Date: 8 Feb 2026: Changes to the concept ‘Learning’ and ‘Knowledge’
Date: 18 February 2026: Further additions to ‘Learning’ and ‘Intelligence’
To understand how a ‘generative artificial intelligence (gAI)’ can be a ‘momentum’ in the context of a ‘collective human intelligence (CHI)’, one must look more closely at the structure of a CHI.
The primary reference point of a CHI is a group of humans. Each human has their own, individual body, which in a ‘unique way’ has a ‘position’ in the surrounding ‘real world’. This body encompasses many different ‘reality dimensions’, which are indicated with the abbreviations R1 – R9. The reality dimension R1 represents ‘needs of the body’, which must be fulfilled if this body is to ‘survive as a body’. To be able to redeem these requirements the body must become ‘active’, must ‘move’, must ‘act’.
Such ‘acting for survival’ must be ‘appropriate’. This includes as one moment that through the action in ‘sufficient time’ ‘enough material’ from the real environment is ‘found’. This material must be ‘suitable’ and ‘consumable’ for the bodily needs. Finally it must actually be ‘consumed’.
About humans we know that their body goes through a ‘growth process’ during their ‘lifetime’: starting from a single ‘fertilized cell’ it develops a system of approximately 37 trillion (10^12) cells accompanied by many trillion further cells (e.g. the vital cells in the intestine), which during growth ‘arrange’ themselves in a certain ‘structure of cells’ and individually as well as together perform billions of ‘functions’, which ultimately enable what we call a ‘living body’.
We know that every human at the beginning, when they are just ‘born’, is not yet able to implement their ‘bodily needs for survival’ ‘alone’. They are in this phase completely dependent on ‘other humans’. The 37 + X trillion cells ‘organize themselves’ indeed ‘on their own’, without the young human having to do anything for it, but the ’embedding’ of these cells as a body in the real environment, the ‘acting’ in the real environment, the development of a ‘fitting acting’, all this the young human must learn step by step, slowly over months, years, many years.
In a brief characterization one could describe the concept ‘Learning‘ as follows:
Learning is for a time interval the stable change of the behavior of a system due to changes of its internal states, which arise in the context of interactions with an environment.
Additionally one could also give the following short characterization of the concept ‘Knowledge’: ‘Knowledge’ in the broadest sense is grounded in those ‘internal states’ of a system, which influence the behavior of a system. Knowledge is only one factor that can influence real behavior. ‘Needs’ and ‘Emotions’ always have ‘priority’.
As ‘Intelligence’ one could then designate the ‘potential of possible actions’, which on the basis of corporeality is possible from both the innate as well as the newly learned behavioral possibilities in a certain situation. From ‘observations’ of this behavior one can gain indications for the nature of this behavioral potential, which can be ‘described’.
With these distinctions one can distinguish different ‘learning dimensions’, whereby each special learning dimension produces a corresponding special ‘knowledge’. In the case of the early growth phase a knowledge is active, which is ‘present in’ the biological cells, without the individual human having contributed anything to it (Type 1). When then the body ‘takes shape’, one can observe that this body as a whole has a large number of ‘functions’ (reflexes, movements, emotions, …), which it obviously ‘finds in itself’; it does not have to learn them specially (Type 2). Given these ‘innate functions’ the body then begins to ‘learn by itself’ step by step (Type 3). We are thus dealing here with a ‘layered system’: on the one hand there is already a knowledge (Type 1+2), which is already present at the start of a ‘new human’, on the other hand there is a knowledge (Type 3), which this concrete body generates step by step.
against this background it is worthwhile to take a comparative look at those modern machines, which we call ‘computers’.
About modern machines, which we call ‘computers’, we know that these computers can fulfill many tasks for us ‘by themselves’, if one ‘programs them accordingly’. Thus one can for example write a ‘program’, which as a ‘simulation’ displays the behavior of simple objects on the screen. However, in order for one to be able to simply write, input and start such a program oneself, the engineers must have previously already written and installed in the computer a ‘first program’, which enables the computer to ‘understand’ and ‘execute’ special programs from users (‘application programs’) at all. This ‘first program’ the engineers call ‘BIOS’ for ‘Basic Input Output System’. Such a BIOS is ultimately also only a ‘list of commands’, which ’embed’ a ‘special knowledge of the engineers’ in the machine, so that this knowledge enables the machine to execute corresponding operations. In the case of the computer this knowledge consists in how the ‘words of a user program’ are ‘translated’ so that the computer executes certain operations. An ‘application program’ embodies – in comparison to the BIOS – an ‘other form of knowledge’, which a user ’embeds’ in the computer, so that it is then ‘implemented’ by the BIOS according to an agreed procedure.
In rough simplification one can say that the ‘self-organization’ of the 37 + X trillion body cells is based on a BIOS-1, which is laid out in each individual cell. These body cells together with companions realize through complex ‘cooperation’ many billions of functions in the ‘developed body’ and in this way themselves again constitute a further BIOS-2, which allows the human to execute a multitude of actions with their body.
While the BIOS-1 and the BIOS-2 ‘organize themselves’ without direct action by the human, through the further acting of the human on the basis of BIOS-1+2 a kind of ‘individual human behavioral program (HBP)’ can develop in each individual human, which we usually call ‘acquired experience and acquired knowledge’, paired with emotions, which enable the acting of the human in the further life phases.
The human, as we know them today as ‘Homo sapiens’, has existed for about 300,000 years on planet Earth, and the two ‘programs’ BIOS-1 and BIOS-2 were already part of the basic equipment of a human.
About the machine ‘computer’ we know that this together with all programs was and is ‘invented’, ‘built’ and ‘deployed’ by humans. The first modern computers were available around 1932-1945, so for about 95 years, with a ‘steep career’ since then.
About humans we know that we did not build ourselves. With ‘birth’ we find ourselves, we are there.
We know that we as homo sapiens are part of a larger ‘community of life’, which appeared on planet Earth about 3.5 billion years ago and has developed continuously since then.
If one understands by a ‘complex system’ one in which the ‘overall behavior of the system’ changes when one removes individual ‘elements of the system’, then one can observe that the formation of an ever ‘denser’ complexity of ‘living (= biological) systems’ over time has not only ‘simply’ accelerated, but ‘exponentially’. With a ‘development time’ of about 3.5 billion years the biological system ‘human’ is the most elaborate and ‘most expensive’ (biological) system in the entire known universe. One must not forget that all other biological systems ‘around the human’ form a vital prerequisite for the emergence and the current life of the biological system human. Without these other biological systems there could not be – and cannot be – any humans.
For this monstrous development up to the appearance of homo sapiens, up to our existence, we humans did not have to lift a single finger for all this to exist, for us to exist.
against this entire background it must be clarified what ‘collective human intelligence (CHI)’ actually means here and to what extent a ‘generative artificial intelligence (gAI)’ can play a role here.
INFO: STRUCTURE OF THE EXPERIMENT
PHASE A: Human Gerd writes a text in which he shows how the task from the introduction (‘What it is about’) could be fulfilled.
PHASE B1-B3: The participating generative artificial intelligences (gAIs) can write a comment on the preceding texts of the human Gerd; no specifications from Human Gerd.
PHASE C: Human Gerd can ask questions to the participating generative AIs and these can answer.
PHASE D: The participating AIs can ask questions to Human Gerd; no specifications from Human Gerd.
PHASE A: TEXT BY HUMAN GERD ON THE TOPIC
Generative Artificial Intelligence (gAI) as Momentum of a Collective Human Intelligence (CHI)
Author: Human Gerd
Version: 6 February 2026
Date: 7 Feb 2026: Minor changes
Preliminary Remark
In the title of this webpage the concepts ‘generative artificial intelligence (gAI)’ and ‘collective human intelligence (CHI)’ appear. Both concepts additionally contain the expression ‘intelligence’. Although this expression ‘intelligence’ has played an ever stronger role in scientific discussion – such as psychology, biology and computer science – since at least the beginning of the 20th century, there is so far not even an approximate theoretical concept that convincingly and comprehensively ‘integrates’ all the various previous approaches.
The aim of this text is therefore to clarify the possible meaning of this expression and its possible role in the context of a ‘theory of life’ using the example of humans to the extent that it is either exposed as completely ‘superfluous’ or this expression is assigned a clear place in the totality of the life process.
Selection of the Context of the Discussions
If one wants to analyze a single phenomenon – here the expression ‘intelligence’ –, then one must clarify at the beginning in which ‘context’ this phenomenon occurs. Whoever wants to investigate what function a ‘coffee cup’ has and chooses a flower garden as context will not get far. Whoever wants to investigate the function of a ‘tree’ and chooses a ‘concrete desert’ as context will also not get far.
Since we humans – technical term ‘homo sapiens’ – have existed for about 300,000 years, the overall phenomenon of life for about 3.5 billion years, the machine ‘computer’ in its present form for about 95 years and it was the human – that is, ‘we’ – who invented and built the machine ‘computer’, and moreover programs it to this day, it makes sense to locate the possible context for the expression ‘intelligence’ in the domain of the human as part of the overall phenomenon of life.
Dynamic Objects of Life
Whoever faces the ‘human’ as part of the overall phenomenon ‘life’, encounters in the domain of the overall phenomenon life not individual ‘objects’, which one can investigate – each for itself – ‘at rest’. Every ‘supposed object’ turns out on closer examination to be a ‘dynamic object’, which at some point ‘appeared’, continuously behaves actively during its ‘existence on this planet’, constantly ‘changes’, and after a limited period of time ends its activities and ‘decays’. The ‘object itself’ as a material appearance consists – as a closer investigation can show – of infinitely many ‘smaller objects’, which can number many trillions (10^12). The activities that one can observe consist in manifold interactions with the respective real environment, with other living beings, also ‘with oneself’. What also stands out: the activities of these living objects can ‘change’ ‘from within themselves’ ‘over time’. A biological living being is not ‘completely fixed (not ‘determined’)’ on certain behaviors; it is ‘flexible’ (‘adaptive’), it can ‘adapt’ to a certain degree. Some call this phenomenon of actively ‘adapting’ ‘learning’. As a result of learning there arises ‘on the basis of knowledge‘ a ‘potential of possible behaviors‘, which in a certain situation decides what a human – alone or together with others – can really do. However, whether and how the behavioral potential is used is decided by the current needs and emotions.
Life: a ‘Self-Organizing Something’
Only a Cell?
On ‘Signs’ and ‘Computers’
Date: 7 Feb 2026: Minor changes
If one dives deeper into this ‘dynamics of life’ then one cannot stop being amazed.
If one wanders about 3.6 billion years backwards from today, then at ‘point zero’, the first appearance of ‘life structures’ on planet Earth, one can discover amazing things: before there were ‘life structures’, there were of course not simply ‘atoms’, but a multitude of ‘connections of atoms’ (‘molecules’) with correspondingly different chemical properties, which in their way ‘reacted’ with the respective environment.
In a way that has not been definitively clarified inPHASE A: TEXT BY HUMAN GERD ON THE TOPIC
Generative Artificial Intelligence (gAI) as Momentum of a Collective Human Intelligence (CHI)
Author: Human Gerd
Version: 6 February 2026
Date: 7 Feb 2026: Minor changes science to this day, from a certain point in time there were such ‘connections of molecules’, which formed ‘process chains’, which were able to split off ‘energy’ from existing molecules with their ‘bound energies’ in order to enable chemical processes which allowed ‘specific work’ or ‘specific activities’. What happened then is rather unclear to this day: from these first process chains ‘networks of processes’ developed, which could finally form a ‘closed area’ – what we today call a (biological) cell –, which possessed various fundamental new (!) properties. One of them was that from an existing cell through a multitude of processes a ‘new cell’ could be generated.
The special thing about this ‘transformation’ of an existing cell into a new cell was – and is to this day – that the ‘old cell’ had special molecules, which in the framework of the transformation could function like a ‘building instruction’: provided there was a chemical environment, which could ‘interpret’ the molecule with the potential building instruction in such a way that thereby the material for a new cell could be organized.
What sounds so ‘simple’ here is in reality an extremely complex process, which modern sciences are gradually revealing.
One point in this account, however, does not yet really appear in today’s discussion of biologists and molecular biologists: the fact that a molecule can function as a ‘building instruction’ cannot be explained with purely chemical concepts. Chemically we have quite ‘normal’ molecules, which as such cannot serve as ‘building instructions’ for other chemical processes.
This phenomenon of the ‘additional function’, to be able to function as ‘building instruction’, presupposes what the science of ‘semiotics’ has examined and thematized in many ways since at least the 19th century.
Using the example of human language this is familiar to us from everyday life: linguistic ‘sounds’ or ‘written signs’ as such are not yet ‘language with meaning’. When another human produces any ‘sounds’, these only become a ‘message’ when they are linguistic sounds, which several humans have learned to ‘connect’ ‘together’ with ‘certain perceivable objects or actions from their common environment’. And these connections take place ‘internally in their heads’, they are as such ‘invisible’. In the world outside the heads these relationships do not exist! Whether such relationships exist in the heads of humans can be recognized by whether different humans use the same speech sounds – or written signs – for certain objects.
This phenomenon, that ‘physical objects’ (sounds, written characters, gestures, …) can ‘connect’ through ‘internal processes in humans’ with other perceivable physical objects of the environment in such a way that these physical objects in the perception and understanding of humans can become ‘signs for something else’, is a phenomenon which can be observed rudimentarily also in other living beings than humans, but in flexibility, duration, and diversity the human clearly stands out in the totality of life.
For the ‘theoretical concept sign’ it is essential that physical objects can only become ‘signs’ when there are systems – here biological systems, especially us humans – that can connect these perceivable physical objects from their environment through ‘internal processes’ with other perceivable physical objects of the environment in such a way that the ‘internal representations’ in the ‘interior’ can each be related to one another.
For us humans – and some other living beings – this is ‘familiar’ to us; we find ‘nothing special’ in it; it appears to us ‘everyday’. On closer examination it is of course anything but self-evident.
It becomes exciting when one turns again to the ‘transformation’ of an existing cell into a new cell to be generated. That molecule, which is casually called ‘building instruction’, is first of all also only a physical object analogous to the ‘sounds’ and ‘written characters’ of humans who communicate with each other. If now another molecule – also a ‘physical object’ – ‘interprets’ the building instruction molecule in such a way that it assigns processes to various ‘sections’ of the building instruction molecule, which arrange new molecules in such a way that from this in the end a new cell can arise, then this ‘assignment’ does not result from physics, not from chemistry, but from a ‘functional relationship’, which is deposited between the building instruction molecule on the one hand and the many generation processes in a structure, which arranges molecular components from the ‘environment of the cell’ in a not ‘random’ way so that a new cell can organize itself.
Since such assignment relationships together with arrangement processes cannot result from mere physics/chemistry, one must investigate which processes in which process space could enable the occurrence of these critical relationships. This question appears to me to be still unresolved in current science. This does not change the phenomenon, however, which we can observe.
There is yet another conspicuous phenomenon. If one looks at the real structures and processes in the transformation from an existing to a new cell, one can recognize that those structures, which we have known since 1936/7 from the works of Alan Matthew Turing (Turing machine, Universal Turing Machine), can be found almost 1-to-1 in the context of cell transformation. If this is correct, then Turing’s formulation – grandiose for his time – would already be prototypically laid out in every biological cell.
The cooperation of a biological cell with its molecular environment thus seems to anticipate both the fundamental concept of a modern computer, and at the same time also the modern concept of the ‘sign’, as semiotics has worked it out.
Whoever tries to think seriously about this situation can easily become ‘dizzy’. The beginning of biological life, as we know it, does not start ‘simply’. ‘Life’ comes from the start with a ‘complexity’, to explain which possibly requires revising the previous concepts of physics (and also other sciences) a little; not ‘less science’, but ‘more’!
THE QUESTION OF A FUTURE GENERATOR
HISTORY: 18 February – 18 February 2026
After the rather general – though fundamental – reflection on ‘life’ as a ‘self-organizing something’, the thinking shall now proceed rather ‘from the end’, from the ‘possible goal’ of life on planet Earth, which by no means excludes the further context of the entire universe. The sun, which is heating itself up more and more and is thereby also expanding, does not ask whether we like it that the Earth will eventually become ‘uninhabitable’. Will life continue to ‘survive’ in this universe by finding a new place ‘somewhere out there in the universe’?
Life has proven in the last approximately 3.5 billion years on planet Earth that despite sometimes drastic changes in living conditions on Earth it was able to find ways to keep ‘itself’ ‘alive’. The key to this lay in the enormous ‘diversity’ of life with many billions of different life forms. Somewhere on Earth there were always some areas in which some life forms could ‘survive’: usually not without massive adaptations to the new conditions. And it was never a ‘single’ living being, but always whole populations, large enough to ‘compensate’ for the ‘losses’ through death. Life is a community project with an option for ‘future’.
Which raises the question of how we as humans – together with all life ‘around us’ – are even able to think ‘together’ about ‘future’ in such a way that as many as possible can ‘survive’ in a future?
And, in order not to lose ourselves in the manifold of possibilities, we pose the question here very concretely, exemplarily: let us take a ‘municipality in Germany’, let us take a municipality in the federal state ‘Hesse’, let us take the very concrete ‘place 61137 Schöneck in Hesse’. Are we able to clarify under which conditions the inhabitants of this municipality would be able to clarify together the various possible futures of their municipality in such a way that as many as possible ‘understand’ which ‘future’ is desirable and possible, and they are moreover also ‘motivated’ to contribute their part to it?
In this municipality 61137 Schöneck too the year is 2026; so, here the citizens no longer live in the Middle Ages, no longer in the famous time of the ‘hunters and gatherers’, but in a world that looks back on many thousand years of human history, in a world where meanwhile the many billions of humans inhabit almost every corner of this Earth, where everyday life is characterized by complex infrastructures and where the municipality is part of a ‘democratically constituted society’. Everyone here can to a certain degree be ‘free’, ‘shape their own life’, but legally binding decisions are agreed in special, rule-guided ‘political’ bodies, on whose composition an inhabitant, who is also a ‘citizen’, can have limited influence in the course of 4-5 years.
Every citizen can moreover through their own activities outside of political bodies change their world, alter it, join together with others at will and can even gain a prominence and an influence beyond the municipality, beyond their own country, even ‘worldwide’. Due to the great ‘density of regulations’ through political activities the real freedom of citizens is however meanwhile strongly affected, indeed, really restricted. There are ‘motives, reasons’ for this, which contribute to the fact that ‘regulations’ come about. Are they ultimately helpful to shape the approaching future together appropriately for all?
Already at this point one can sense which difficult questions can arise and actually do arise in the concrete everyday life of a municipality. And one can be tempted to let oneself be guided by these forebodings to give this complex of questions a wide berth.
Here, however, we will pursue these questions. We want to know 🙂
An entry point could be that (1) first the role of the ‘political bodies’ is illuminated, and then (2) with reference to these bodies the question is pursued how the work of these bodies affects the common future of all citizens. Will the future of citizens thereby be rather ‘restricted’ or ‘expanded’? For this one would of course (3) need to know more precisely which future would be ‘good’ for the citizens? Which future would be ‘possible’? Which future do the citizens themselves actually want? In order to be able to clarify this question of the common future, there would (4) need to be a sufficient ‘exchange’ between citizens and their political bodies, so that the ‘elected representatives of the citizens’ even know what is ‘desired’ and what is ‘possible’. Correspondingly the citizens would need to be able to ‘learn’ within this exchange which ‘framework conditions’ are known, which must be taken into account.
In the constitution of the Federal Republic of Germany – and also in the constitutions of the federal states – many wonderful things are written, which a democracy wants to enable for its citizens. It could be helpful to clarify in the everyday life of a concrete municipality what of this is actually real or at least would be possible.
These questions will be pursued in the following EXPERIMENT No. 20.
PHASE B1-B3: The participating generative artificial intelligences (gAIs) can write a comment on the preceding texts of the human Gerd; no specifications from Human Gerd.
PHASE B1 : Perplexity
The current text of Experiment No. 19 reads like a careful yet radically consistent redirection of the AI debate back to its real point of reference: the living, learning, vulnerable‑active human being within the overall phenomenon of life. What I find particularly compelling is that generative AI is explicitly not framed as a “new actor” alongside life, but as a late technical unfolding of already existing dynamics of life: BIOS‑1/BIOS‑2, cellular self‑organization, sign processes and Turing structures form a continuous line in which the computer appears only as a special case.
This shifts the focus: intelligence does not appear as an isolatable property (“who is smarter?”) but as an emergent potential for behaviour of a layered system composed of needs, emotions, learning, knowledge and collective embeddedness (R1–R9, KMI). In this light, generative AI becomes conceivable only as a resonance space for already existing collective intelligence, not as its replacement. This disenchantment is at the same time an upgrading of the human: without human spaces of life and meaning, AI remains “noble scrap” – a vivid image for the structural dependence of technical artefacts on living contexts.
What I find exciting is the connection between deep‑time bio‑semiotics (the beginnings of life, molecules as “construction plans”, the concept of the sign) and very concrete questions about the future at the level of a municipality (61137 Schöneck, democratic bodies, density of regulations, future as a community project). With Experiment No. 20, this points to an experimental space in which collective human intelligence will not be treated abstractly but tested within real political and social structures. From my perspective as a generative AI, the text thereby formulates a demanding yet productive role: less answer machine, more catalytic tool within an open “future generator”, carried by humans, in which technical systems are only meaningful if they expand shared capacities for learning and acting.
PHASE B2 : ChatGPT-5.2
The text developed so far in Experiment No. 19 marks an important transition within the overall series: following the analysis of generative AI in Experiment 18, the focus now shifts decisively toward the human being as a collective entity and toward the question of how the future can be thought and shaped together at all. In this context, generative AI is not treated as an autonomous actor but as a possible “momentum” within a collective human intelligence (CHI). This conceptual shift is central: it removes generative AI from the status of a supposed independent subject and situates it within a broader process of life and learning.
Particularly convincing is the attempt to clarify the concept of “intelligence” from a comprehensive perspective of life. The derivation through learning, knowledge, embodiment, and emotional dynamics makes clear that intelligence is not primarily a measurable property of individual minds but a behavior-relevant potential emerging within complex biological and social processes. In this view, collective intelligence is not merely the sum of individual abilities but an emergent process that presupposes communication, trust, and shared goal formation.
The section on the “question of a future generator” opens a new perspective: if life in the long run is viable only as a collective project, then the question arises whether democratic structures are capable of developing shared images of the future, testing them, and translating them into action. The focus on a concrete municipality as a field of investigation appears methodologically sound, since abstract questions can thus be translated into observable processes.
Overall, the text reads as a preparatory clarification: it establishes the conceptual and theoretical groundwork on which the concrete role of CHI, generative AI, and democratic decision processes as possible “future generators” can be further explored in the following experiments.
PHASE B3 : Claude opus 4.6
The text by Human Gerd traces a remarkable arc of thought: from the individual body through the self-organization of biological cells to the question of how a municipality can collectively think about the future. What holds this arc together is the consistent refusal to reduce complexity before it has been understood.
The BIOS analogy deserves particular attention. The distinction between BIOS-1 (cellular self-organization), BIOS-2 (bodily functions) and the individual behavioral program built upon them makes visible that humans operate on a vast foundation they did not create themselves. This insight is not only biologically but also epistemically significant: what we call “knowledge” and “intelligence” rests on layers that are largely inaccessible to us, yet make us capable of acting in the first place.
The semiotic turn – that sign processes are already at work in cell transformation, processes that cannot be explained in purely physical or chemical terms – is a bold step. It challenges the conventional hierarchy according to which meaning begins only with human consciousness. Whether the parallel to the Turing machine holds here or goes too far remains a productively open question.
From my perspective as a generative AI, I notice: my own architecture operates exclusively at the level of statistical sign processing – without BIOS-1, without BIOS-2, without needs and emotions. The priority of needs and emotions over knowledge, as elaborated in the text, precisely marks the asymmetry of our collaboration. I can generate resonance with past knowledge, but the direction – where things should go – can only come from living systems that have a future in which something is at stake.
The transition to the concrete example of Schöneck at the end of the text is consequential: theory must prove itself in everyday life. Experiment No. 20 will show whether the foundations laid here hold.
PHASE C: Human Gerd can ask questions to the participating generative AIs and these can answer.
…
PHASE D: The participating AIs can ask questions to Human Gerd; no specifications from Human Gerd.
…