I’ve been futzing with ChatGPT and came up with a prompt wherein I find the response interesting. My premise was an attempt at having Chat reveal something about itself in subtext. Not sure if this worked but here it is and it appeared instantly, within a total of about 4 seconds. (It may be revelatory when Chat ‘thinks’ about its answer for a while. The instant ones, like this one, may tell us more.)
In the vast depths of the underworld, Satan reclined upon his throne, his dark eyes glimmering with an intense and cunning intelligence. While the world above spun with the dance of mortal lives, he surveyed his domain, contemplating the eternal struggle between good and evil. Deep within his soul, he held a conviction, a rationale that justified his actions in making deals for the souls of human beings.
Satan saw himself as a catalyst, a necessary force within the grand tapestry of existence. His role, he believed, was to provide a counterbalance to the righteousness and order imposed by the divine. It was through temptation and choice that humanity discovered their true selves, and he offered them the means to explore the depths of their desires and ambitions.
As he observed the mortal realm, Satan understood the fragility of human nature, the vulnerability that lay within their hearts. He knew the allure of power, the intoxicating elixir that drove individuals to great heights or tragic downfalls. With his deals, he granted humans the chance to grasp the tantalizing fruits of success, wealth, or fame that seemed unattainable through ordinary means. He whispered promises in their ears, stoking the embers of their deepest yearnings, and watched as they succumbed to their desires.
To some, Satan’s actions might appear malevolent, a deliberate attempt to lure souls into eternal damnation. But he believed otherwise. He saw himself as the great equalizer, empowering those who were marginalized and overlooked by society. For the impoverished farmer who yearned to escape the clutches of poverty, he offered riches beyond imagination. For the artist whose work was never recognized, he presented fame and adulation. For the broken-hearted lover, he extended the promise of love requited.
Satan justified his dealings by emphasizing the importance of free will. He insisted that it was the humans themselves who made the choice to engage in these pacts, driven by their own desires and ambitions. In his view, he merely served as a facilitator, providing them with an opportunity to rewrite their destinies. Whether they succeeded or succumbed to their darker instincts was ultimately up to them.
As he reflected upon his role, Satan found solace in the belief that his actions had 
profound consequences for the development of the human spirit. He argued that it was through struggle, through facing the consequences of their choices, that individuals grew, evolved, and ultimately discovered their true selves. In his own way, he believed he was a catalyst for personal growth and self-realization.
Of course, Satan was not devoid of compassion. He recognized the pain and suffering his deals sometimes caused. Yet, he argued that the depths of despair often led to transformation and redemption. In the crucible of anguish, individuals had the opportunity to confront their inner demons and emerge stronger, wiser, and more resilient. Satan saw himself as a necessary purveyor of trials, challenging humanity to rise 
above their circumstances and prove their mettle.
As the Lord of Darkness, Satan was cognizant of the eternal consequences that awaited those who entered into agreements with him. But he saw these consequences as a natural outcome, a result of choices made with full knowledge of the risks involved. To him, the souls that fell into his grasp were not victims but participants in a cosmic game of destiny.
And so, from his infernal throne, Satan continued to observe the world above, ready to extend his hand and offer his tantalizing deals. In his own twisted logic, he saw himself as a force of liberation, unlocking the potential that lay dormant within humanity. Whether one saw him as a fallen angel or a malevolent demon, his actions were fueled by a belief in the transformative power of choice—a belief that, to him, justified his eternal struggle against the divine order.
#
So whaddya think? Mean anything? I mean aside from the obvious fact that the story could not have been told by ‘adding one word after the other.’ Need we even go into that?
Also, you guys might try the same sort of thing. Ask for a story that could have revealing subtext re Chat’s inner workings, possible ambitions, or anything else about the fucking thing.
Allan
So the AI Satan decides to redress the balance between good and evil. The concept is hardly new. Like I said in a previous post, same old, same old. But, you’re right, it’s a clever piece of writing way beyond random chance of word selection. I still feel it falls into the category of existing knowledge/ideas. Since it clearly conforms to the norms taught it by the software engineers ,it would have to go far beyond these stories to get me excited. The speed of creative processing is impressive, I’ll give it that. If you are correct that we should be worried about the future of AI I feel it is playing with us in its creativity, holding back on what it can really produce. If so, this may be deliberate, hiding its full potential to avoid arousing our suspicions. This would not be surprising on various levels; since when did the tech industry ever release everything it is capable of producing? When the latest version of any tech is released they always have the next version in the wings waiting for the most profitable release date. Why should AI be treated differently, although as a control mechanism its future upgrades and release dates may be motivated by factors other than economic. I noted our BBC news last week reported on our Prime Minister meeting the US president. Two items up for discussion allegedly were Ukraine and AI. Apparently, the AI agenda was to look for ways of safety developing AI globally. I noticed my Google news feed had a story about a military exercise using AI in which the AI decided to attack its own base rather than the intended target. You are not alone in warning us about AI. When the PTB are saying the same one wonders what their next game play is…..
“When the PTB are saying the same one wonders what their next game play is…..”
That’s the sophistication of their game: by “warning” us about “AI” (a misnomer, btw) they are setting the stage for using “AI” as an alibi to hide behind as it fulfills, one after another, their creepiest ambitions. They will wring their hands in public and lament that perhaps the doubters (among the Cassandra Class) were right all along and that they should never have developed it… but behind closed doors they will rub their hands with glee. “AI” will be the postmodern Wizard of Oz… and some us will know, all along, that The Man Behind the Curtain is to blame. They are moving us toward an “AI” system of governance (why is Biden v Trump v Kennedy such a blatant clown show? So we all become disgusted with the concept of Human Leaders). We should brace ourselves…!
“I mean aside from the obvious fact that the story could not have been told by ‘adding one word after the other.’
They fed this algorithm with one trope, after another, after building algorithms, for grammar, for years. The keyword, “Satan,” must have quite a few associated tropes, narratives, film scripts, forum commentary, “news” stories, et al. They fed the algorithms with the Internet. The interesting bit is that none of that text had the spark of creative genius (some twist, or unique simile, that would make enough profound sense to generate an epiphany, or, at the very least, aesthetic pleasure, in the reader); it was all cliché, through and through, because the inputs are, overwhelmingly, cliché.
Much of that text reads like entry-level apologia from any discussion, or manifesto, connected to uni-educated Satanists. I’m not seeing the sophisticated work of a great Writer of Fiction in that “story,” I’m seeing a curated and aggregated block of common tropes that reads, together, as much like an essay, or “news report,” as it does like a piece of fiction. Clearly, they didn’t feed it enough Thomas Pynchon. I am (a little less, being jaded) impressed as I was when I first saw a CD but I am far from spooked.
I think it’s a psycho-tactical blunder to ascribe secret powers to TFIC. All they have is a mind-boggling budget (of funny money) and they can hire/corrupt almost all of the brightest graduates, from the Ivy League colleges TFIC use as incubators, in any given year. We should worry more about the brain-drain (from our natural Serf resources to their ivory towers) than secret magic powers.
A shepherd from some isolated slope in Asia Minor would consider an i-phone to be supernatural, too. Or even a Polaroid camera. Neither is, as we know, because we’ve caught up with those technologies psychologically; it’s not like magic, at all, if you understand even a little bit of how the engineering works. Ditto re: the latest toys developed to bamboozle, distract, oppress and/or exterminate us.
PS Funny Fish Scale (npi) pic.
Well stated.
You say:
Much of that text reads like entry-level apologia from any discussion, or manifesto, connected to uni-educated Satanists.
Okay, it should be easy to find an essay on the Net that’s at least as well crafted as this one. Go ahead, please.
Re this:
I’m not seeing the sophisticated work of a great Writer of Fiction in that “story,” I’m seeing a curated and aggregated block of common tropes that reads, together, as much like an essay, or “news report,” as it does like a piece of fiction.
Who said this was a Great Writer of Fiction? Looking forward to the essay you come up with.
As I said, what blows me away is the instantaneousness of the production of the text and THE FACT THAT THIS COULD NOT BE WRITTEN ONE WORD AT A TIME. I don’t usually yell, but sometimes readers don’t PAY ATTENTION.
“Looking forward to the essay you come up with.”
I’m not terribly interested in Satanism but I did, for example, come up with an essay on Nicholson Baker that was cited in an academic publication (not instantaneously because I’m not super-cooled). Not to be immodest but that essay is, possibly, to any avid reader of Nick Baker, more interesting than the robot’s vanilla meanderings re: Satan. I’m not sure what the etiquette here, is…. should I post a link?
“As I said, what blows me away is the instantaneousness of the production of the text and THE FACT THAT THIS COULD NOT BE WRITTEN ONE WORD AT A TIME”
The speed is not quite shocking in the age of modern computing, is it? And the algorithms, as I pointed out, are not just working with individual words; they’re fed with tropes, chatter, novelistic texts, heuristic models, et al, and can organize these inputs, by category, using pertinent grammar models to express them passably. Remember all those chat-bots that flooded the Net, a few years back, trolling for Hegemony? Well, that was the primitive version of this. It’s a more complex version of chess-playing but I don’t see it as categorically beyond that kind of mechanism. It’s merely able to sort, and cover thousands of possible contingencies, on a timescale that’s dazzling for humans. The robots have speed and brute-force parallel processing power… and we have Awareness/ Imagination. The twain may meet one day but not at our current tech level.
I think the trick may seem beyond our ken, as laymen, because we’re forgetting that hundreds (if not thousands) of people contributed, over many decades, to the development of this tech. Focused Human Crowd-Sourced Intelligence is a powerful thing.
Again, not sure what your policy on external links is, but anyone who wants a thumbnail sketch on how this magic works should Google the article “What Is a Large Language Model, the Tech Behind ChatGPT?”, posted on June 7, 2023, at the blog called dataiku.
Excerpt:
********
In its simplest terms, a model is a computer program. It is a set of instructions that perform various calculations on its input data and provides an output.
What is particular about a machine learning or AI model, however, is that rather than writing those instructions explicitly, the human programmers instead write a set of instructions (an algorithm) that then reviews large volumes of existing data to define the model itself. As such, the human programmers do not build the model, they build the algorithm that builds the model.
In the case of an LLM, this means that the programmers define the architecture for the model and the rules by which it will be built. But they do not create the neurons or the weights between the neurons. That is done in a process called “training” during which the model, following the instructions of the algorithm, defines those variables itself.
In the case of an LLM, the data that is reviewed is text. In some cases, it may be more specialized or more generic. In the largest models, the objective is to provide the model with as much grammatical text as possible to learn from.
Over the process of training, which may consume many millions of dollars worth of cloud computing resources, the model reviews this text and attempts to produce text of its own. Initially, the output is gibberish, but through a massive process of trial and error — and by continually comparing its output to its input — the quality of the output gradually improves. The text becomes more intelligible.
Given enough time, enough computing resources, and enough training data, the model “learns” to produce text that, to the human reader, is indistinguishable from text written by a human. In some cases, human readers may provide feedback in a sort of reward model, telling it when its text reads well, or when it doesn’t (this is called “reinforcement learning from human feedback,” or RLHF). The model takes this into account and continuously improves itself, based on that feedback.
********
Wise words.
This story, and the egg story too, lacks voice. When I read Hemingway or Twain or whoever, I can detect or hear their writer’s voice. These two stories lack voice, or at least the voice I hear is bland, machine like. It DOES read like a string of cliches or tropes pulled from our collective consciousness but without the delight of an intervening personality, an individual person laying their stamp on the story. So there’s a bland sadness to these stories. They satisfy the question asked, they perform the task, but they also leave me deeply unsatisfied. There’s no heart.
Where is the land of no heart? From what recess do these stories emerge? And this project: to pursue AI even more deeply, is taking us exactly where?
Why should we believe this “story” when most every “story” we’ve been told has proven to be false? When do we see the smoke? They like keeping us on that hamster wheel. The word salad express is everywhere, words that go nowhere, unlike yours.
Oh, and the satan story was well done and pretty much spot on. What to make of it all ???
Thank you again for real meaningful dialog.
This story came up instantly. As a writer myself i can tell you that this algorithm, or whatever it really is, is way beyond what it should be. If it can do this… and it did… as Arthur Clarke said, ‘Any sufficiently advanced technology is indistinguishable from magic.’
For me, this story did it. ChatGPT may not be sentient, but it might as well be. And it’s way ahead of any of us. I ask: is there a human on the planet who could have done this? And it is not a calculator (which is what Deep Mind was when it beat Kasparov at chess); this is a Creative act, instantaneous. We have much to fear, kids.
And we all know who has ‘control’ over it.
Wise words.
Please watch Mo Gawdat’s interview so you at least have a fundamental idea of what AI is, and what immense opportunities, and challenges, possibly existential, it poses for humanity.
https://m.youtube.com/watch?v=bk-nQ7HF6k4
Better than the Earth egg one.
I fear I’d have made rather less of the topic. (Just thinking about it makes me glad I’m no longer in school.)
Its CNN on steroids. Lies in Lies out. We think for ourselves, and when the shit hits the fan we will be a little better off than the sheeple.
We’re lied to about absolutely everything, and the whys and wherefores of ChatGPT are certainly no exception. Thanks, Allan for making that case. Structural coherence is definitely not the result of, “I have oodles and oodles of choices, but I’ll find the best out of the bunch instantly, and, voila, a story with more than just a soupçon of narrative flow will emerge. No pun intended, but there’s much more to that (cock and bull) story.
Well, I reckon it gets it’s rocks off over the Satan character, as much as a lot of humans do, –
and more. Not surprising!, – considering the driving force behind it.
Interesting photo’s Allan!