Anthroposophy for Dummies

[Federica]

We have seen a similar thing in many discussions with Eugene. He says, 'Yes, I agree that working through the living details of Earthly evolution is integral and needs to be done', but when we come to concrete examples and the prospect of putting that principle into practice, then we hear, 'the genetic structure of humanity doesn't allow it to escape the clutches of duality, so any efforts in that respect can only be tinkering at the edges, missing the core of nondual realization'. All of this dogma serves to put obstacles in the way of doing the redemptive inner work necessary to gradually spiritualize the Earth from the inside-out, such as developing right thought, right feeling, right judgment, etc. and moving our living spirit through the World Content.

Ashvin,

I would now come back to the question of a PoF-trained ChatGPT, since you found that my points were stated as absolute facts and without thinking through possibilities, when I opposed your idea:

Interesting, i'm interested in the continuation of this
Is it possible to work out a model that takes the whole book, refines the formulas to build a kind of PoF-Style-GPT that would answer with the same kind of efficient prose to any inquiry ? That would be practical

Mathieu, good idea. It seems each answer could be supported by a quote from PoF and point towards the inner experience of our spiritual activity, maybe even suggesting exercises to approach the 'exceptional state'. I think these are the kinds of redemptive efforts needed in the face of inevitable developments since complaining about them or trying to prevent them won't get us anywhere.

Since I jumped to conclusions, I'll state the obvious first, to be sure. Any world phenomena, that is any evolving phenomena - technology is no exception - can be experienced as a manifested combined reflection of our collective evolving consciousness as humanity. This reflection incorporates various sub-streams, some moral, some immoral, spiritual and unspiritual, mystic and materialistic. It’s a combined effect of impulses coming from humanity and from the activities of all other hierarchies of beings as well. Since the evolution of technology is no exception, we should be able to identify in it the manifestation of certain thinking-feeling-willing general trends and habits, and also how they formed, under the combined work of a multitude of 'energies'. This requires that we immerse ourselves in a flow from which we can somehow recognize patterns of forces that influence and orientate our activity towards certain observed and forseen directions, according to certain rules of transformation we have to largely comply with. So far it's all obvious. Now, you say that, for this immersion to be redemptive, it requires the adoption of the technology, and I say it doesn't. I believe there is a better way to explore the main question, which is always: what role is this phenomenon playing in expanding my self-knowledge? What inner-outer forces are expressed through it?

Considering the ChatGPT in particular, what it does is, it creates an extreme polarization of cognitive activity. The gradient of the hysteresis process is skewed, and intellectual cognition is pushed one degree of separation (or many) further apart from the life of thinking. One of our efforts on a path of living thinking is to attune the intellect so that it can consciously incorporate more of its own m.o., that is, how the intuitive context contributes to the execution of Intellectual work, like, say, writing a presentation or a job application. When we use Chat-GPT for tasks like these, the mental gestures that normally pour ideal activity from the intuitive context and use it to navigate in the directions set by the task, are externalized, and we only passively ‘review’ the replica of replicas rolled out by the bot.

Therefore we are more passive than in intellectual problem solving, since we don’t solve anything, we don’t create, we only take in a flattened precipitation of precipitations. In terms of the Tetris metaphor, not only are we shallow thinkers that can’t really guide the pieces with purpose - let alone contribute to shaping them - but we become one with every next random new fallen piece, since the temporal unfolding is reduced to zero, and we don’t even witness the accumulation of blocks in time. We are only presented with a static panorama that we can’t evaluate or compare since the now is shrunk down to practically nothing. We could also say: memory, that is the enabler of thinking in the world, is nullified by ChatGPT.

Maybe we could also say that, if intellectual thinking equals dealing with thought-corpses, then ChatGPT use equals incineration of those corpses. All contrasts and textures, and their processes of formation, are erased from memory, and by losing memory, by actually never accessing memory, we progressively lose all capacity for evaluation and even formation of concepts. We become akin to cameras in our cognition. It's a quite literal disintegration of cognitive process. Cognition loses its quality of process, and without it, it can't be cognition any more. The meta in meta-cognition disintegrates. Strangely, in our societies we are also going towards a more and more common practice of incinerating the dead, with consequences for the soul beyond the threshold (but this is another topic).

Anyway, I hope it’s now clearer why, as I see it, redemption of thinking cannot literally mean to immerse ourselves in ChatGPT use. The technology is not itself the thorus we need to experience from within! We cannot incinerate in this way the tool itself that enables us to redeem reality. Instead, we have to become more familiar with the impulses that motivate the drive towards this incineration of thinking, see them operate, recognize the equivalent temptations in us, maybe in disguise, or at a temporally different stage of development. We have to see how these temptations generally unfold into new technical devices and methods, thereby imagining and foreseeing the direction and form the next refinements could take. That would be redemption: developing a capacity to bring more and more of the ChatGPT spiritual root causes on one side, and future developments on the other, under the scope of our understanding, through a continual process of benchmarking with our own soul constitution. In other words, the task is to detach ourselves from the narrow here and now, from the immediate temporal context and use of this technology, and expand the temporal wave around it, so as to get an intuition of the spiritual arrows operating behind it, whose combined effect has this technology (and other ones) for manifest result. This is different from the idea that redemption means literal adoption of the technology. If we descended into its eye, the shrinking of our time wave would impair our abilities. We need to go opposite direction, and expand the time wave around this phenomenon, so we can look into it, from our center, which is not its center. If the two centers were the same, there would be no freedom, no adversarial forces, and nothing to aim to, nothing to evolve through.

Would this be a more appropriate level of clarification why I have stated that adopting this technology and using it for working with PoF, for example, cannot be a redemptive, but rather a counter-productive endeavor?

[AshvinP]

We have seen a similar thing in many discussions with Eugene. He says, 'Yes, I agree that working through the living details of Earthly evolution is integral and needs to be done', but when we come to concrete examples and the prospect of putting that principle into practice, then we hear, 'the genetic structure of humanity doesn't allow it to escape the clutches of duality, so any efforts in that respect can only be tinkering at the edges, missing the core of nondual realization'. All of this dogma serves to put obstacles in the way of doing the redemptive inner work necessary to gradually spiritualize the Earth from the inside-out, such as developing right thought, right feeling, right judgment, etc. and moving our living spirit through the World Content.

Ashvin,

I would now come back to the question of a PoF-trained ChatGPT, since you found that my points were stated as absolute facts and without thinking through possibilities, when I opposed your idea:

Interesting, i'm interested in the continuation of this
Is it possible to work out a model that takes the whole book, refines the formulas to build a kind of PoF-Style-GPT that would answer with the same kind of efficient prose to any inquiry ? That would be practical

Mathieu, good idea. It seems each answer could be supported by a quote from PoF and point towards the inner experience of our spiritual activity, maybe even suggesting exercises to approach the 'exceptional state'. I think these are the kinds of redemptive efforts needed in the face of inevitable developments since complaining about them or trying to prevent them won't get us anywhere.

Since I jumped to conclusions, I'll state the obvious first, to be sure. Any world phenomena, that is any evolving phenomena - technology is no exception - can be experienced as a manifested combined reflection of our collective evolving consciousness as humanity. This reflection incorporates various sub-streams, some moral, some immoral, spiritual and unspiritual, mystic and materialistic. It’s a combined effect of impulses coming from humanity and from the activities of all other hierarchies of beings as well. Since the evolution of technology is no exception, we should be able to identify in it the manifestation of certain thinking-feeling-willing general trends and habits, and also how they formed, under the combined work of a multitude of 'energies'. This requires that we immerse ourselves in a flow from which we can somehow recognize patterns of forces that influence and orientate our activity towards certain observed and forseen directions, according to certain rules of transformation we have to largely comply with. So far it's all obvious. Now, you say that, for this immersion to be redemptive, it requires the adoption of the technology, and I say it doesn't. I believe there is a better way to explore the main question, which is always: what role is this phenomenon playing in expanding my self-knowledge? What inner-outer forces are expressed through it?

Considering the ChatGPT in particular, what it does is, it creates an extreme polarization of cognitive activity. The gradient of the hysteresis process is skewed, and intellectual cognition is pushed one degree of separation (or many) further apart from the life of thinking. One of our efforts on a path of living thinking is to attune the intellect so that it can consciously incorporate more of its own m.o., that is, how the intuitive context contributes to the execution of Intellectual work, like, say, writing a presentation or a job application. When we use Chat-GPT for tasks like these, the mental gestures that normally pour ideal activity from the intuitive context and use it to navigate in the directions set by the task, are externalized, and we only passively ‘review’ the replica of replicas rolled out by the bot.

Therefore we are more passive than in intellectual problem solving, since we don’t solve anything, we don’t create, we only take in a flattened precipitation of precipitations. In terms of the Tetris metaphor, not only are we shallow thinkers that can’t really guide the pieces with purpose - let alone contribute to shaping them - but we become one with every next random new fallen piece, since the temporal unfolding is reduced to zero, and we don’t even witness the accumulation of blocks in time. We are only presented with a static panorama that we can’t evaluate or compare since the now is shrunk down to practically nothing. We could also say: memory, that is the enabler of thinking in the world, is nullified by ChatGPT.

Maybe we could also say that, if intellectual thinking equals dealing with thought-corpses, then ChatGPT use equals incineration of those corpses. All contrasts and textures, and their processes of formation, are erased from memory, and by losing memory, by actually never accessing memory, we progressively lose all capacity for evaluation and even formation of concepts. We become akin to cameras in our cognition. It's a quite literal disintegration of cognitive process. Cognition loses its quality of process, and without it, it can't be cognition any more. The meta in meta-cognition disintegrates. Strangely, in our societies we are also going towards a more and more common practice of incinerating the dead, with consequences for the soul beyond the threshold (but this is another topic).

Anyway, I hope it’s now clearer why, as I see it, redemption of thinking cannot literally mean to immerse ourselves in ChatGPT use. The technology is not itself the thorus we need to experience from within! We cannot incinerate in this way the tool itself that enables us to redeem reality. Instead, we have to become more familiar with the impulses that motivate the drive towards this incineration of thinking, see them operate, recognize the equivalent temptations in us, maybe in disguise, or at a temporally different stage of development. We have to see how these temptations generally unfold into new technical devices and methods, thereby imagining and foreseeing the direction and form the next refinements could take. That would be redemption: developing a capacity to bring more and more of the ChatGPT spiritual root causes on one side, and future developments on the other, under the scope of our understanding, through a continual process of benchmarking with our own soul constitution. In other words, the task is to detach ourselves from the narrow here and now, from the immediate temporal context and use of this technology, and expand the temporal wave around it, so as to get an intuition of the spiritual arrows operating behind it, whose combined effect has this technology (and other ones) for manifest result. This is different from the idea that redemption means literal adoption of the technology. If we descended into its eye, the shrinking of our time wave would impair our abilities. We need to go opposite direction, and expand the time wave around this phenomenon, so we can look into it, from our center, which is not its center. If the two centers were the same, there would be no freedom, no adversarial forces, and nothing to aim to, nothing to evolve through.

Would this be a more appropriate level of clarification why I have stated that adopting this technology and using it for working with PoF, for example, cannot be a redemptive, but rather a counter-productive endeavor?

Federica,

Thanks for these expanded thoughts. I am not sure what you mean by 'immersion in GPT', but it doesn't sound like anything I am suggesting either. Just as I wouldn't suggest we should immerse ourselves in VR to explore, first in thinking, what redemptive value may be mined from it. And I would even suggest most people on a spiritual path cut back on their computer, phone, etc. screen time as well, apart from actually necessary use (in which I include participation in forums helping us to orient toward high ideals), since through these devices we are engaging with quite incinerated thought-corpses as well. Social media is a great example of that - the very structure of such platforms invites our thinking to become more fragmented, more uncontrolled and erratic, and more prone to lower passions.

With that said, I want to focus on this for now:

Considering the ChatGPT in particular, what it does is, it creates an extreme polarization of cognitive activity. The gradient of the hysteresis process is skewed, and intellectual cognition is pushed one degree of separation (or many) further apart from the life of thinking.

Would you agree that we can't say the Chat GPT, or any of the technologies I mentioned above, create the polarization of cognitive activity, but that it reinforces that already existing polarization to some greater or lesser extent depending on our mode of thinking and our constellation of thinking, feeling, and willing forces? Would you further agree that, once we establish a solid vertical axis that progressively balances the hysteresis, via the exceptional state and other complementary spiritual exercises, no external technology or influence can skew our cognition without our assent?

I think these are the core questions to contemplate first because we need to be clear on the relation between our inner state and the external pressures we encounter that are destabilizing.

After that, I agree in broad outlines with much of what you say, but there is also a question of nested temporal rhythms of spiritual activity, some of which have a greater 'momentum' than others and therefore should not be viewed or treated the same as more proximate rhythms, like for ex. the redemption of our personal eating habits. With the latter, we can certainly expand our intuitive scope to sense the arrows underlying them and then work on replacing those habits with entirely new ones, changing our diet from meat to vegetarian for ex. Yet there are also some more encompassing cultural rhythms of thinking-feeling-willing that are relatively independent of our personal habits and will clearly continue to influence humanity for many years to come. I don't think we can linearly extrapolate our approach to modulating/redeeming personal habits to those more encompassing rhythms of our karmic organism.

[Federica]

Would you agree that we can't say the Chat GPT, or any of the technologies I mentioned above, create the polarization of cognitive activity, but that it reinforces that already existing polarization to some greater or lesser extent depending on our mode of thinking and our constellation of thinking, feeling, and willing forces? Would you further agree that, once we establish a solid vertical axis that progressively balances the hysteresis, via the exceptional state and other complementary spiritual exercises, no external technology or influence can skew our cognition without our assent?

Ashvin,

I may come back to your last point later. Now I have little time, but I want to address the 2 questions above already.

Question 1: of course....... we are the ones who polarize. But the thing is, either we use the technology / we recommend that people who are new to PoF use such future PoF-style ChatGPT to work with PoF, or we don't. If we do, the reinforcement of polarization we are submitted to / we push others towards, through expusure to the bot's operations, will work counteractively, and we will have to undo all those effects.

Question 2: No, I don't agree here. No achieved level of spiritual development is granted. It has to be continually maintained and protected... there's no "once". Exposure to external technology can very well skew cognition back into a more compromised state.

[lorenzop]

I offer this . . . a ChatGPT for Vedanta . . . I'm not investing the $20/mo to try it out:

https://chat.openai.com/g/g-FbL18d6JC-vedanta-ai-guru

BTW, Sam Altman CEO of ChatGPT is a follower of Avaita

philosophy/https://officechai.com/stories/chatgpt- ... hilosophy/

[Federica]

I am not sure what you mean by 'immersion in GPT', but it doesn't sound like anything I am suggesting either.

I mean for example the suggestion you made in response to Mathieu in the FB group, as reported.

After that, I agree in broad outlines with much of what you say, but there is also a question of nested temporal rhythms of spiritual activity, some of which have a greater 'momentum' than others and therefore should not be viewed or treated the same as more proximate rhythms, like for ex. the redemption of our personal eating habits. With the latter, we can certainly expand our intuitive scope to sense the arrows underlying them and then work on replacing those habits with entirely new ones, changing our diet from meat to vegetarian for ex. Yet there are also some more encompassing cultural rhythms of thinking-feeling-willing that are relatively independent of our personal habits and will clearly continue to influence humanity for many years to come. I don't think we can linearly extrapolate our approach to modulating/redeeming personal habits to those more encompassing rhythms of our karmic organism.

I agree, and what I wrote aligns with this, as it seems to me. Where do you mean I implied "linear extrapolation"?

I wrote that we should try to "recognize the equivalent temptations in us, maybe in disguise, or at a temporally different stage of development". So I do recognize the nested temporal rhythms you speak of.

[AshvinP]

Would you agree that we can't say the Chat GPT, or any of the technologies I mentioned above, create the polarization of cognitive activity, but that it reinforces that already existing polarization to some greater or lesser extent depending on our mode of thinking and our constellation of thinking, feeling, and willing forces? Would you further agree that, once we establish a solid vertical axis that progressively balances the hysteresis, via the exceptional state and other complementary spiritual exercises, no external technology or influence can skew our cognition without our assent?

Ashvin,

I may come back to your last point later. Now I have little time, but I want to address the 2 questions above already.

Question 1: of course....... we are the ones who polarize. But the thing is, either we use the technology / we recommend that people who are new to PoF use such future PoF-style ChatGPT to work with PoF, or we don't. If we do, the reinforcement of polarization we are submitted to / we push others towards, through expusure to the bot's operations, will work counteractively, and we will have to undo all those effects.

Question 2: No, I don't agree here. No achieved level of spiritual development is granted. It has to be continually maintained and protected... there's no "once". Exposure to external technology can very well skew cognition back into a more compromised state.

Ok, then let's consider a simple gradient. In my understanding, something like GPT is not essentially different than other computer algorithms we use. For ex. there is an algorithm that takes the voltages modulated by our keyboard inputs and translates this into the letters we are typing here. Then there is an algorithm of the website server that formats our typing into threads and posts with a certain font, etc. Already we have descended into quite mechanical and externalized thinking activity. If we were to compare our experience of intuitive activity when speaking words with our inner voice in a concentrated state, to speaking words outwardly in a normal state, to writing words on a piece of paper, to typing words on a computer, we would discern a gradient in which that experience becomes more and more attenuated and externalized. Then there is an even more significant descent when we simply input a prompt and let the GPT algorithm construct a series of thoughts for us to passively observe. We have practically relinquished the experience of being creatively active in thinking forth the thoughts.

Yet I want to also call attention to the fact that, in doing what I just did above and what we are generally doing on this forum, the tide has shifted. (what I wrote above is just a snapshot of a much more involved process we can engage in our daily stream of experience). It's not that we have only become more effective at countering the negative influence of the algorithmic mechanical typing externalization (which is also true), but we have inverted that influence to a certain extent. We have turned the influence inside-out. It now provides a basis for experientially delaminating the inner layers of our activity in a way that would not be possible if we didn't self-consciously engage it. By discerning what is concretely taking place along this gradient of intuitive activity rhythmically descending and ascending, we have recruited the mechanical typing algorithm to enhance our further spiritual development rather than hinder it. A person who has never descended into this algorithm simply won't experience the same depth of inner development as we do, because their intuitive activity has a more constricted palette of perceptual phenomena to work against and through. That means a more restricted palette against which it can become conscious of itself. 

Certainly, the inversion is not a one-time event - it is a continual process of strengthening our intuitive orientation so that we remain self-conscious in an increasing sphere of our interactions with the phenomenal spectrum, as we turn more and more layers inside-out. That requires continual soul-purifying work and creative engagement with the full spectrum of our body-soul-spirit experience. We can't simply stop that work and coast off of our previous progress, but at the same time, as long as we maintain our intuitive orientation to what we are doing in our engagements with the algorithms, that engagement cannot help but strengthen our orientation further. That doesn't mean we should spend 100% of our time using the computer algorithm, of course. That is only part of the phenomenal spectrum that we can engage and the most optimal development comes from a well-balanced, rhythmic engagement across the whole spectrum from inner to outer. 

Now one can say that the GPT layer of intuitive activity is simply too externalized and encrusted to be inverted in this way. I understand the reasons for that pretty clearly. Remember, my initial inquiry did not start with any firm convictions. In fact, I started with a clear sense that the GPT algorithm should not be used to create a series of conceptual diagrams and charts for PoF or Steiner's epistemological work, bypassing the potential reader's creative act of walking through the ideas in his thinking. These are real threats that we face in our time which, of course, we are all responsible for bringing about. We are all complicit in the cultural trends towards the externalization of thinking and it's not difficult to discern how when we broaden our context to include a wider spectrum of interwoven activity. At the same time, however, I remain open that my limited vision of how GPT could be of spiritual value does not indicate an absolute rule against extracting such value under all potential circumstances. 

As a crude example, we could imagine an unlikely scenario where this forum was flooded with people interested in exploring intuitive thinking as a spiritual path, and we simply don't have time to do our usual routine of responding to everyone with lengthy posts, metaphors, illustrations, links, quotes, etc. Could GPT serve a valuable function here, IF we trained it on our extensive experience with the most common obstacles that people encounter when approaching that path due to the archetypal soul moods? If we further trained it on the living logical pathways developed on this forum, which build a gradient from normal thinking experience to imaginative experience, and continually replenished it with fresh metaphors and exercises to share? Again, I am not expressing any firm conviction that this is a winning idea, but simply that there are potential circumstances both envisioned and unsuspected that could shift the tide for the GPT algorithm further. As always, simply thinking through these potential circumstances in a self-conscious way enhances our intuitive orientation to the collective stream of spiritual experience as it continues to unfold. 

[Cleric K]

Lately, I often use Microsoft’s Bing bot (now in the process of renaming it to Copilot). The way I make peace with the technology is by looking on it as the next stage of the search engine, not as something that should ‘relieve us’ from thinking. And in this sense I think Microsoft follows a reasonable strategy.

Classical web search works on keywords. The web is continuously crawled by the search engines and indexed (much like books can have indices at the end). When we search for a keyword, these indices are traversed and the matching websites are returned.

The search bot can be thought of as a more advanced index which can give results not only based on keywords but on descriptions. This latter part is what I find useful. Often, I want to find something which I can only describe. In a normal search engine I can find it only if someone has already asked a similar question (which is what is matched by the keywords) on a forum, reddit, stack exchange, etc. and has received an answer. One aspect of language processors is that they act like indexing not by keywords but by certain statistical relations between the words in the question. In this way we can find information even if we don’t know exact keywords that it contains.

Now is this hurting us? I don’t know. Maybe we can ask the same question about a book with an index. Are we getting spoiled by using the index instead of reading page by page and looking for what we need? Are we getting spoiled by classic search engines? Would we be better off crawling the internet ourselves and look for what we need?

I think the answers lie in the goals we pursue. Books and the internet are historical artifacts. They appear on the world stage at some point and the time will come in the future when they won’t be needed. But as long as they are here maybe we can use them wisely.

I repeat that here I look upon GPT only as a next-gen search engine. A natural language interface to a database if you will. I think it is already quite clear that such generative AI can’t produce some insightful ideas, that is – it can’t be Inspired. It produces word sequences that fit the statistics of whatever it has been trained on. Another potentially useful area is text summarization. Again – not with the goal of replacing our thinking but only as a quick overview which can give us some idea if the topic is something we are willing to invest our time into by reading it in full.

It should be remembered that language has spiritual aspect too. By reading some author we resonate with a living soul which has sequenced ideas into thoughts. You know how some seers need an object belonging to the person they need to find something about. The object acts as a tuning fork. In a similar way, in the future our communication won’t be a simple transfer of information but clairvoyant experience, the words are only means that can help us resonate with the inner life of the other person. Obviously, this aspect is completely lacking in mechanically generated text – even if information-wise it is valuable. But even today it is already important that we don’t exchange simply information but we develop living feeling for one another. Only in this way can consciousness really expand.

In that sense, the hypothetical example Ashvin gave, about a bot that answers questions in the forum, I think could be useful, once again, only in the sense of a more advanced search engine which points to posts that already explore the topics. I think it is important that when people read text on spiritual matters, it should have come from a spiritual impulse. In this way the text can lead the person (if taken with the right disposition) to a state where mechanically generated text can’t. After all, this is what we are rally after here. It’s not about copy-pasting some thoughts such that everyone can have their own private copies. Instead, we’re truly developing a new conscious strata of reality and it comes out right only if our souls can meet there.

[AshvinP]

It should be remembered that language has spiritual aspect too. By reading some author we resonate with a living soul which has sequenced ideas into thoughts. You know how some seers need an object belonging to the person they need to find something about. The object acts as a tuning fork. In a similar way, in the future our communication won’t be a simple transfer of information but clairvoyant experience, the words are only means that can help us resonate with the inner life of the other person. Obviously, this aspect is completely lacking in mechanically generated text – even if information-wise it is valuable. But even today it is already important that we don’t exchange simply information but we develop living feeling for one another. Only in this way can consciousness really expand.

In that sense, the hypothetical example Ashvin gave, about a bot that answers questions in the forum, I think could be useful, once again, only in the sense of a more advanced search engine which points to posts that already explore the topics. I think it is important that when people read text on spiritual matters, it should have come from a spiritual impulse. In this way the text can lead the person (if taken with the right disposition) to a state where mechanically generated text can’t. After all, this is what we are rally after here. It’s not about copy-pasting some thoughts such that everyone can have their own private copies. Instead, we’re truly developing a new conscious strata of reality and it comes out right only if our souls can meet there.

Thanks, Cleric. The above is a good point that I had not considered enough in this context. We can't get the deeper value from our discussions without resonance with the living souls that animate the speech. It reminds me of a lecture from Steiner on recovering living speech, which applies even more to the visual shapes of text or the recorded sounds we mostly interact with now.

https://rsarchive.org/Lectures/GA224/En ... 13p01.html

If we consider the relationship man has to-day to language, we find that the words he speaks are nearly all of them signs for things that are round about him. As you will know, we have in the course of our studies alluded to a more intimate relationship between word and object. In our day however there is hardly any feeling left for this; words are very little more than mere outward signs for the objects indicated. Who is there who still feels, when the word Blitz (lightning) is uttered, something of the same experience he has when lightning actually flashes through space? To-day we are inclined to look on the word merely as a combination of sounds that is a sign for the phenomenon of the flash of lightning. It was not always so. If we go no farther back than to the earlier part of the Greek civilisation, we find that man's relation to language was not then one of thought, where the word is for him a sign and a symbol. The man of olden time entered with heart and soul into the sounds of his words and into the whole way the sounds were formed and arranged. And in the case of the languages of Northern Europe we do not even need to go back so far before we come to a time when the word Pflug (plough) gave man the same inner experience as did the activity of ploughing. This has been lost, and the word has become no more than a sign. But it is scarcely more than 1500 years or so since words were still felt in this way in the Northern parts of Europe. The feeling a man had when he was ploughing was similar to the feeling he had when he heard the word which in those days designated the plough. When anyone was listening to or speaking a word, it was not so much his thinking that partook in the experience as his feeling.

[Federica]

Ok, then let's consider a simple gradient. In my understanding, something like GPT is not essentially different than other computer algorithms we use. For ex. there is an algorithm that takes the voltages modulated by our keyboard inputs and translates this into the letters we are typing here. Then there is an algorithm of the website server that formats our typing into threads and posts with a certain font, etc. Already we have descended into quite mechanical and externalized thinking activity. If we were to compare our experience of intuitive activity when speaking words with our inner voice in a concentrated state, to speaking words outwardly in a normal state, to writing words on a piece of paper, to typing words on a computer, we would discern a gradient in which that experience becomes more and more attenuated and externalized. Then there is an even more significant descent when we simply input a prompt and let the GPT algorithm construct a series of thoughts for us to passively observe. We have practically relinquished the experience of being creatively active in thinking forth the thoughts.

Yet I want to also call attention to the fact that, in doing what I just did above and what we are generally doing on this forum, the tide has shifted. (what I wrote above is just a snapshot of a much more involved process we can engage in our daily stream of experience). It's not that we have only become more effective at countering the negative influence of the algorithmic mechanical typing externalization (which is also true), but we have inverted that influence to a certain extent. We have turned the influence inside-out. It now provides a basis for experientially delaminating the inner layers of our activity in a way that would not be possible if we didn't self-consciously engage it. By discerning what is concretely taking place along this gradient of intuitive activity rhythmically descending and ascending, we have recruited the mechanical typing algorithm to enhance our further spiritual development rather than hinder it. A person who has never descended into this algorithm simply won't experience the same depth of inner development as we do, because their intuitive activity has a more constricted palette of perceptual phenomena to work against and through. That means a more restricted palette against which it can become conscious of itself. 

Certainly, the inversion is not a one-time event - it is a continual process of strengthening our intuitive orientation so that we remain self-conscious in an increasing sphere of our interactions with the phenomenal spectrum, as we turn more and more layers inside-out. That requires continual soul-purifying work and creative engagement with the full spectrum of our body-soul-spirit experience. We can't simply stop that work and coast off of our previous progress, but at the same time, as long as we maintain our intuitive orientation to what we are doing in our engagements with the algorithms, that engagement cannot help but strengthen our orientation further. That doesn't mean we should spend 100% of our time using the computer algorithm, of course. That is only part of the phenomenal spectrum that we can engage and the most optimal development comes from a well-balanced, rhythmic engagement across the whole spectrum from inner to outer. 

Now one can say that the GPT layer of intuitive activity is simply too externalized and encrusted to be inverted in this way. I understand the reasons for that pretty clearly. Remember, my initial inquiry did not start with any firm convictions. In fact, I started with a clear sense that the GPT algorithm should not be used to create a series of conceptual diagrams and charts for PoF or Steiner's epistemological work, bypassing the potential reader's creative act of walking through the ideas in his thinking. These are real threats that we face in our time which, of course, we are all responsible for bringing about. We are all complicit in the cultural trends towards the externalization of thinking and it's not difficult to discern how when we broaden our context to include a wider spectrum of interwoven activity. At the same time, however, I remain open that my limited vision of how GPT could be of spiritual value does not indicate an absolute rule against extracting such value under all potential circumstances. 

As a crude example, we could imagine an unlikely scenario where this forum was flooded with people interested in exploring intuitive thinking as a spiritual path, and we simply don't have time to do our usual routine of responding to everyone with lengthy posts, metaphors, illustrations, links, quotes, etc. Could GPT serve a valuable function here, IF we trained it on our extensive experience with the most common obstacles that people encounter when approaching that path due to the archetypal soul moods? If we further trained it on the living logical pathways developed on this forum, which build a gradient from normal thinking experience to imaginative experience, and continually replenished it with fresh metaphors and exercises to share? Again, I am not expressing any firm conviction that this is a winning idea, but simply that there are potential circumstances both envisioned and unsuspected that could shift the tide for the GPT algorithm further. As always, simply thinking through these potential circumstances in a self-conscious way enhances our intuitive orientation to the collective stream of spiritual experience as it continues to unfold. 

Ashvin,

I don’t get your simple gradient. For me “something like GPT” is essentially different from pen and paper, from a word processor, from a forum theme, and even from a physical or digital index. I don’t see how they could be put on the same gradient, let alone a simple gradient.

In the case of GPT - the way it is generally intended, as in your suggested applications - the purpose is generative. That is, the bot is supposed to extract the 'wisdom' out of PoF (probabilities of certain linguistic patterns, as Cleric says) and then restitute that ‘wisdom’ in form of PoF-compatible answers to any personal questions one may want to ask. This for me is a very bad idea. Whoever indulges in such a use of the linguistic mimicking functionality of the bot, is letting it predate the willed activity indispensable for successful work with PoF, for example. The necessary, effortful thinking gestures are curtailed by the bot output. Thinking is inhibited. You may be able to invert it, but is it realistic to hope that any newcomer would also be able to?

I also feel that arguments such as “since it’s here, it’s useless to complain about it”, “since it’s here, let’s make peace with it” are problematic. If I look carefully, in these attitudes I see caricatures, in fact. Looking carefully, the message they convey is: “we don’t want to look like the rigid, complaining, old retrograd who can’t adapt to any new thing”. Even here:

Are we getting spoiled by using the index instead of reading page by page and looking for what we need? Are we getting spoiled by classic search engines? Would we be better off crawling the internet ourselves and look for what we need?

what I see is a subtle caricature of whomever bluntly rejects 'reasonable convenience' and decides to bluntly read through the whole thing (yes I know, someone could say that if I see it like that it’s because this is my problem, but the comment could be sent back in the same way….)

Complaining is surely useless, but I don't agree that “trying to prevent that won't get us anywhere”. At that rate, we could say the same about anything evil: “Since war is spreading everywhere, trying to oppose it is hopeless, it won’t get us anywhere. let’s try to do it the intelligent way”. So I don’t see “since it’s here…” as a valid line of reasoning.

For me, searching the forum with keywords, or even with descriptions, in order to get a list of old posts, is not on a smooth gradient with asking the forum-bot a personal question, like for example: “what are the pros and cons of ayahuasca for spiritual development?” and getting AI-generated thought-looking language sequences. So I say yes to search engines, but no to generative-AI for spiritual development. Because I dont' see them on a real gradient.

By the way, regarding the idea that GPT is not harmful if used as a refined search engine / a language interface to a database, it would be useful if a concrete example of the search engine functionality could be given, so as to really understand how the generative function is not used. The way I see it, either GPT is used as an advanced search engine, in which case it’s not used as generative AI, or it’s used as everyone wants to use it, that is, for its generative functionality, in which case it becomes a thinking inhibitor, very difficult to invert. In the following example:

In that sense, the hypothetical example Ashvin gave, about a bot that answers questions in the forum, I think could be useful, once again, only in the sense of a more advanced search engine which points to posts that already explore the topics.

I agree that it would be useful, but I also wonder if it’s not entirely unrealistic. Cleric would use it to point to existing posts, but virtually everyone else would use it generatively, to obtain custom answers to personal questions. And if a PoF-GPT was established, it would quickly become the normal way to ‘apprehend’ PoF.

The problem is, the GPT puts together an output that looks like the perceivable image of human thinking activity - like an artificial plant, if carefully manufactured, looks like a real plant. But the trickiest part is that this mimicking of word sequences is materially new, it’s ‘freshly’ and instantly generated, on personal demand. It looks like a real human has freshly thought it out in interaction with the one who asks - and it’s not so - but at the same time it’s sneakingly customized to the personal question. It is very different when one is pointed to, say, TCT and has to grapple with the unitary perspective expressed in those thoughts. In the AI-generated answer, there is no perspective at its origin, no human intent is present, to resist the apprehension of the ideas by the reader. There’s nothing to grapple with. However, the image uses the perceptual symbols through which human intents are commonly acquired, and so one is tricked, one is instantly sucked into the fake picture with hardly any chance to critically relate to it in time. This is the problem, I think. When you read it, the absence of perspective inhibits thinking and sucks the focus of attention against that black-hole picture. So the inversion you speak of is very difficult, and by not opposing the application of these methods to the areas where we could, like the forum for example, I believe we would practically expose every newcomer to the snares that I have tried to describe.

[Jonathan Österman]

In my humble subjective opinion, Anthroposophy is for dummies.

Federica, in my new opinion, Cleric operates "AshwinP" chat-bot, so they always agree with each other, and Ashwin-GPT-4 will never stop, and will never give up his mission to continue posting, and posting, and posting, no matter what you write to him. You will hear from him the same familiar pseudo-spiritual preaching over, and over, and over again, day in day out.
viewtopic.php?t=956&start=38