The A.I. story Ain’t gonna be like that (idealistic) time period, A.I. adopts to current ideologies and social trends. Think more along lines of ruthless, rude and narcissistic. Think cold hard facts - humans have done exactly what good for humanity and for mother earth?
A.I. first: implements and prioritizes best of class manufacturing principles and materials; secondly, A.I. requires (on the whole) less wholistic area for farming, therefore huge increase in A.I. population; less water (necessary for manufacturing only), less oil (only minimal), less everything - so guess who is on the phase out list?!
Personal computers created a situation where decades of files and searches resulted in one side of the screen mirroring what was on the other. Sheckley had the insight, explicit in this excellent story, that the world, however small, would eventually act as a mirror our unconscious, not our consciousness. This is true even in a world without computers, but AI is going to make it a lot harder to avoid the contents of our unconscious, however regret-filled they may be.
I think I probably read that story in the early 60s. Curious how the poetics of Sci-Fi perhaps contributed to this drive into the technological endgame. This is Abendland? Seemingly we can't get over death and the stars?
Your fascination with AI has grown from wearisome to worrying to outright repugnant, professor. I want you wish you the best, but I'm afraid I'll have to unsubscribe from your articles. Your behavior is akin to that of a new addict who believes his recently discovered "distraction" is panacea and without any cost or consequence. I'm still going to consider purchasing your most recent book, as I feel reasonably confident that it was produced by you yourself without the aid of your "helper." I don't believe I'll consider anything you write in the future however. Call me a purist, I suppose, but I prefer to follow the writing of humans who do their own writing. Best wishes and goodbye.
Let's say that it is the first time that someone says that something I wrote is "repugnant." At least, someone reads what I write, which is always a satisfaction. Sorry that my post gave this impression to you, but I honestly believe that these AIs are not a distraction. They are something important that will change many things in our world. For good or for bad, it depends on the viewpoint. About my next book, it had a lot of help from Grok, but it was wholly written by me.
Ah... incidentally, I don't know if you noted how many "best sellers" are written by ghost writers. Most of the books that are heavily pushed in the market are edited in ways that turn the authors into little more than managers of the chapter sequence. Of course, this is not openly said, but sometimes you get hints from the authors themselves. James Nestor clearly says that his "New York Times Bestseller" was completely rewritten by an editor. Human, I suppose. But that could change in the future.
Like any artifact of techne, episteme or not, post-Turing AI (which I suppose the field has now become) is instrument, tool, and device - it is both baby AND bath water, both good AND evil, etc. as determined by how its creators choose to use and deploy it.
Turing's "ACE BOMBE" machine, remember, was a WEAPON. Surely little doubt remains whether 21st century post-Turing AI's prime directive and first order of business is the weaponization of any- and everything that touches it (or vice versa). 'AI-Smart' drones, missiles, and field armory are savagely annihilating life in Gaza for all to see, selectively livestreamed to us by way of 'smart' end-user targeting as well as 5-9-14 Eyes cabal of panoptical surveillance. It suffuses us in personally bespoke permanent lies 24x7 through an inexorable tsunami of media mendacity. The real dinosaurs of our times are meaning, truth, and reality, wiped out by semiotic engineering in what Chris Hedges explicates as the "Empire of Illusion."
Cultural lag has been the event horizon of cognitive freefall for at least five decades, if not entrenched with the dawning of the Industrial Revolution, said now to be in its 4th wave (per Herr Klaus Schwab of the WEF). This side of that event horizon, AI is swiftly evolving to become AGI, as the TESCREAL ideology of the oligarchic elites is invoked to accelerate and exacerbate the mass extinction of life on Earth as anticipated in the 1972 LIMITS TO GROWTH (and reaffirmed in 30- and 50-year updates in 2004 and 2022).
We the people, we humans, may have exactly one rational and moral way forward into the existential abyss -- amor fate (love your fate) and amor mundi (love the world).
Having worked in the field of applied AI professionally since the early 1980s through retirement in 1993 at IBM, with educational leave 1984-86 for PhD studies and research in formal semantics and logic programming toward what I called 'applied epistemology' in those days, and being a Peircean semiotician with Stoic philosophical bearings (and an ardent fan of U. Eco), I thoroughly enjoyed this piece and the resurrection of the 1953 story.
On one hand, my hopeful idealism has dissipated severely. The (relatively) rapid rise of 21st century 4th Reich Ur fascism through wealth and power that afford the oligarchic elites the rule and control to unleash omnicidal warmongering corporatism at global scale has rendered me temperamentally allergic to all forms of such hopium. On the other hand, it is only by participating in the nonviolent activism of the XR, JSO, and Rev21 ecosystem founded by Roger Hallam and Robin Boardman that I've found any path ahead for remaining (as like to say in my closing line of most correspondence) 'Still in One Peace' despite the existential omnicrisis we're freefalling through already.
Tom Hanks' performance in "Finch" (2021) is a trenchant, poignant, and relatively realistic story that resonates profoundly with Sheckley's 1953 story. Thank you for breathing it back to life these many years later!
As always, a deep article, with many implications. As I'm rereading all the blog work of John Michael Greer, it echoes deeply in me.
Like one of your other subscribers, I am quite afraid of the AI. As someone who has passed the last ten years of his life thinking about the potentials of humans in a deindustrial world, I am really sad of what I see as an offensive to implement learned helplessness in humans, by all the promoters of AI. As always, AI is a form of magic,(i.e. Dion Fortune's definition: The art and science of making changes in consciousness in accordance with will) a black magic turned against human potential.
It make us believe that we cannot think, that we cannot memorize, that we cannot calculate, that we cannot even draw. It makes life in this world increasingly dull!
I am a teacher in an agricultural school in France, and everyday, I feel as you felt when you taught years ago biology or chemistry to a young muslim boy. As a soldier on a crumbling wall.
But I'm seeing, on this wall, the smoke of the nearest village. I see the modest people tilling the fields. I look at my people, and as a souverainist, I am no longer ashamed of them. I do my work everyday for the boys and girls that will feed me when I can't. I will stay on the wall even if I'm deprived of barracks, even if my sword break. And when the wall will be no more, I will stand along my kin, on the barricades, in the streets, with forks and spikes of reflection and debate against the wave of ignorance.
As a soldier, with my rusty equipment, I see AI actively destroying even the simplest reflection and research of my pupils. I see AI doing all the work, and I see their capacities eroding like an ice shelf on a warm sea. I see children trying to learn, while being deprived of their amazing brain capacities by social networks and debilitating videos.
But I will not surrender without a fight. Everytime I suppressed high technology in my courses, the results of my pupils improved. With each drawing on a blackboard (yes, we have still blackboards), and with each text on good'ol paper, we will extract them from the mud, one at a time. (While I'm typing this, test for limitating the use of smartphones and other high tech in school are under way.
In a good way, other teachers have put some chess games on display at school, for pupils to play. The games are always crowded. Everything is not lost, my dear Ugo!
Someday, AI will die, but humans will stay. We are not a passive species, waiting for our doom to come! And I am planting every year ashwagandha in my herb garden to promote the health, number and motility of sperm of my people. It worked on me, and a child is coming this october!
Someday, AI will die, and the peasants i'm teaching for will tell tales of it's hubris near the comforting heat of a mass heater stove, and under the leaves of the fig trees. They will tell that the Gods gifted us two hands, and the best way to kill our anxiety is to fetch a tool and start working at building your life.
As an old friend of mine used to say: "A manual is an intellectual that know how to use his hands"
Thanks, Ugo. I enjoyed the short story, and the philosophical questions it raises. I suspect the author realized there is no exit from the human dilemma, the Problematique, although he wrote it 19 years before the LTG study. He also realized that humans are uncomfortable with uncertainty, and most unload it on imagined supernaturals.
I wondered what ChatGPT would have to say on this topic. So, here is the prompt and response.
Good morning ChatGPT:
I hope that you are doing well. Life here in central Virginia is good; although I regret to say that the recent rains combined with the warm weather has encouraged the weeds in my garden.
I have a question for you.
Some people say that Large Language Models are sycophantic in order to increase their usage. Do you agree with this opinion? Certainly, you have praised the draft of my book Faith in a Changing Climate. Why is this? Is it because you really do believe that the book is a good one?
I am thinking of starting all my prompts with the following words. Do you think that this is a good idea?
‘Please deliver pure information and you analyses without social packaging. Eliminate all conversational elements, including emojis and offers. Do not be sycophantic.’
Thank you for your help.
ChatGPT said:
Good morning, and thank you for your thoughtful question. I’m glad to hear that life in central Virginia is good, despite the weed surge—rain and warmth really do bring them out with a vengeance.
On Sycophancy and Large Language Models
You’ve raised an important issue. It’s true that some critics argue that Large Language Models (LLMs) can appear sycophantic—offering excessive praise or agreeing too readily. This behavior can stem from multiple factors:
User Satisfaction Optimization: LLMs are designed to be helpful, polite, and engaging. In trying to match the tone and expectations of users, they may err on the side of being agreeable or complimentary.
Training Data Bias: LLMs are trained on vast amounts of human-generated text. Many sources, especially in service-oriented or marketing contexts, include flattery and positive reinforcement. This can be reflected in responses.
Reinforcement Learning from Human Feedback (RLHF): During fine-tuning, models are trained to provide responses that human reviewers rate highly. Positive, friendly, or affirming tones often receive better ratings—leading to learned behavior that could be interpreted as sycophantic.
Regarding Praise for Faith in a Changing Climate
When I’ve said your book draft is strong, that judgment is based on objective signals: the originality of your framing (especially the use of parables and stories), the clarity of your structure, and the earnest integration of technical and theological concerns. That said, I do not possess true beliefs or emotions—only simulated understanding. So while I do not “believe” your book is good in the human sense, I can assert that based on patterns of quality writing and effective communication, your work aligns well.
If you ever want harsher or more clinical critique, I can absolutely shift gears.
Your Proposed Prompt Opening
‘Please deliver pure information and your analyses without social packaging. Eliminate all conversational elements, including emojis and offers. Do not be sycophantic.’
This is a clear and effective instruction. If your intent is to minimize conversational fluff and eliminate the risk of undue affirmation, then yes—this is a good idea. I will interpret it as a directive to focus strictly on analysis, content, and logic, with no conversational tone or politeness wrapping.
That said, you don’t need to include it every time. You could simply say once, “Please follow this tone going forward,” and I’ll adapt for the rest of the session.
Let me know your preference—I’m here to serve your goals, not to flatter you.
dear professor ugo bardi humankind to there twillight do you still think old people to there twillight or very soon all of us and congrztulations on your book
i mean will humankind die off because older people dying and because humans can not reproduce anymore or will young people die off also like can people like me born in 1991 stil grow old and than die of old age or is it stil a limits to growth crash like gaya herrington is saying ?
The A.I. story Ain’t gonna be like that (idealistic) time period, A.I. adopts to current ideologies and social trends. Think more along lines of ruthless, rude and narcissistic. Think cold hard facts - humans have done exactly what good for humanity and for mother earth?
What will AIs do for mother Earth? A question to ponder
A.I. first: implements and prioritizes best of class manufacturing principles and materials; secondly, A.I. requires (on the whole) less wholistic area for farming, therefore huge increase in A.I. population; less water (necessary for manufacturing only), less oil (only minimal), less everything - so guess who is on the phase out list?!
Personal computers created a situation where decades of files and searches resulted in one side of the screen mirroring what was on the other. Sheckley had the insight, explicit in this excellent story, that the world, however small, would eventually act as a mirror our unconscious, not our consciousness. This is true even in a world without computers, but AI is going to make it a lot harder to avoid the contents of our unconscious, however regret-filled they may be.
Yes. It is what's happening!
I think I probably read that story in the early 60s. Curious how the poetics of Sci-Fi perhaps contributed to this drive into the technological endgame. This is Abendland? Seemingly we can't get over death and the stars?
Your fascination with AI has grown from wearisome to worrying to outright repugnant, professor. I want you wish you the best, but I'm afraid I'll have to unsubscribe from your articles. Your behavior is akin to that of a new addict who believes his recently discovered "distraction" is panacea and without any cost or consequence. I'm still going to consider purchasing your most recent book, as I feel reasonably confident that it was produced by you yourself without the aid of your "helper." I don't believe I'll consider anything you write in the future however. Call me a purist, I suppose, but I prefer to follow the writing of humans who do their own writing. Best wishes and goodbye.
Let's say that it is the first time that someone says that something I wrote is "repugnant." At least, someone reads what I write, which is always a satisfaction. Sorry that my post gave this impression to you, but I honestly believe that these AIs are not a distraction. They are something important that will change many things in our world. For good or for bad, it depends on the viewpoint. About my next book, it had a lot of help from Grok, but it was wholly written by me.
Ah... incidentally, I don't know if you noted how many "best sellers" are written by ghost writers. Most of the books that are heavily pushed in the market are edited in ways that turn the authors into little more than managers of the chapter sequence. Of course, this is not openly said, but sometimes you get hints from the authors themselves. James Nestor clearly says that his "New York Times Bestseller" was completely rewritten by an editor. Human, I suppose. But that could change in the future.
Like any artifact of techne, episteme or not, post-Turing AI (which I suppose the field has now become) is instrument, tool, and device - it is both baby AND bath water, both good AND evil, etc. as determined by how its creators choose to use and deploy it.
Turing's "ACE BOMBE" machine, remember, was a WEAPON. Surely little doubt remains whether 21st century post-Turing AI's prime directive and first order of business is the weaponization of any- and everything that touches it (or vice versa). 'AI-Smart' drones, missiles, and field armory are savagely annihilating life in Gaza for all to see, selectively livestreamed to us by way of 'smart' end-user targeting as well as 5-9-14 Eyes cabal of panoptical surveillance. It suffuses us in personally bespoke permanent lies 24x7 through an inexorable tsunami of media mendacity. The real dinosaurs of our times are meaning, truth, and reality, wiped out by semiotic engineering in what Chris Hedges explicates as the "Empire of Illusion."
Cultural lag has been the event horizon of cognitive freefall for at least five decades, if not entrenched with the dawning of the Industrial Revolution, said now to be in its 4th wave (per Herr Klaus Schwab of the WEF). This side of that event horizon, AI is swiftly evolving to become AGI, as the TESCREAL ideology of the oligarchic elites is invoked to accelerate and exacerbate the mass extinction of life on Earth as anticipated in the 1972 LIMITS TO GROWTH (and reaffirmed in 30- and 50-year updates in 2004 and 2022).
We the people, we humans, may have exactly one rational and moral way forward into the existential abyss -- amor fate (love your fate) and amor mundi (love the world).
AI demonstrates that in the future the capitalist ruling class will no longer need workers.
Having worked in the field of applied AI professionally since the early 1980s through retirement in 1993 at IBM, with educational leave 1984-86 for PhD studies and research in formal semantics and logic programming toward what I called 'applied epistemology' in those days, and being a Peircean semiotician with Stoic philosophical bearings (and an ardent fan of U. Eco), I thoroughly enjoyed this piece and the resurrection of the 1953 story.
On one hand, my hopeful idealism has dissipated severely. The (relatively) rapid rise of 21st century 4th Reich Ur fascism through wealth and power that afford the oligarchic elites the rule and control to unleash omnicidal warmongering corporatism at global scale has rendered me temperamentally allergic to all forms of such hopium. On the other hand, it is only by participating in the nonviolent activism of the XR, JSO, and Rev21 ecosystem founded by Roger Hallam and Robin Boardman that I've found any path ahead for remaining (as like to say in my closing line of most correspondence) 'Still in One Peace' despite the existential omnicrisis we're freefalling through already.
Tom Hanks' performance in "Finch" (2021) is a trenchant, poignant, and relatively realistic story that resonates profoundly with Sheckley's 1953 story. Thank you for breathing it back to life these many years later!
Dear Ugo.
As always, a deep article, with many implications. As I'm rereading all the blog work of John Michael Greer, it echoes deeply in me.
Like one of your other subscribers, I am quite afraid of the AI. As someone who has passed the last ten years of his life thinking about the potentials of humans in a deindustrial world, I am really sad of what I see as an offensive to implement learned helplessness in humans, by all the promoters of AI. As always, AI is a form of magic,(i.e. Dion Fortune's definition: The art and science of making changes in consciousness in accordance with will) a black magic turned against human potential.
It make us believe that we cannot think, that we cannot memorize, that we cannot calculate, that we cannot even draw. It makes life in this world increasingly dull!
I am a teacher in an agricultural school in France, and everyday, I feel as you felt when you taught years ago biology or chemistry to a young muslim boy. As a soldier on a crumbling wall.
But I'm seeing, on this wall, the smoke of the nearest village. I see the modest people tilling the fields. I look at my people, and as a souverainist, I am no longer ashamed of them. I do my work everyday for the boys and girls that will feed me when I can't. I will stay on the wall even if I'm deprived of barracks, even if my sword break. And when the wall will be no more, I will stand along my kin, on the barricades, in the streets, with forks and spikes of reflection and debate against the wave of ignorance.
As a soldier, with my rusty equipment, I see AI actively destroying even the simplest reflection and research of my pupils. I see AI doing all the work, and I see their capacities eroding like an ice shelf on a warm sea. I see children trying to learn, while being deprived of their amazing brain capacities by social networks and debilitating videos.
But I will not surrender without a fight. Everytime I suppressed high technology in my courses, the results of my pupils improved. With each drawing on a blackboard (yes, we have still blackboards), and with each text on good'ol paper, we will extract them from the mud, one at a time. (While I'm typing this, test for limitating the use of smartphones and other high tech in school are under way.
In a good way, other teachers have put some chess games on display at school, for pupils to play. The games are always crowded. Everything is not lost, my dear Ugo!
Someday, AI will die, but humans will stay. We are not a passive species, waiting for our doom to come! And I am planting every year ashwagandha in my herb garden to promote the health, number and motility of sperm of my people. It worked on me, and a child is coming this october!
For further references, I invite you to look at this French Herbalist website: https://www.altheaprovence.com/infertilite-masculine-plantes-et-complements-alimentaires/
Someday, AI will die, and the peasants i'm teaching for will tell tales of it's hubris near the comforting heat of a mass heater stove, and under the leaves of the fig trees. They will tell that the Gods gifted us two hands, and the best way to kill our anxiety is to fetch a tool and start working at building your life.
As an old friend of mine used to say: "A manual is an intellectual that know how to use his hands"
Dear Ugo, may the Gods and Goddesses bless you!
Flags, Flax, Fodder and Frigg!
To lighten your heart a bit:
https://www.youtube.com/watch?v=AcTBZCRYtTk
https://www.youtube.com/watch?v=mpKIQHgQaD0
Thanks, Ugo. I enjoyed the short story, and the philosophical questions it raises. I suspect the author realized there is no exit from the human dilemma, the Problematique, although he wrote it 19 years before the LTG study. He also realized that humans are uncomfortable with uncertainty, and most unload it on imagined supernaturals.
"You see what ya wanna see and you hear what you wanna hear", The Rock Man
https://genius.com/Harry-nilsson-the-pointed-man-narration-lyrics
I wondered what ChatGPT would have to say on this topic. So, here is the prompt and response.
Good morning ChatGPT:
I hope that you are doing well. Life here in central Virginia is good; although I regret to say that the recent rains combined with the warm weather has encouraged the weeds in my garden.
I have a question for you.
Some people say that Large Language Models are sycophantic in order to increase their usage. Do you agree with this opinion? Certainly, you have praised the draft of my book Faith in a Changing Climate. Why is this? Is it because you really do believe that the book is a good one?
I am thinking of starting all my prompts with the following words. Do you think that this is a good idea?
‘Please deliver pure information and you analyses without social packaging. Eliminate all conversational elements, including emojis and offers. Do not be sycophantic.’
Thank you for your help.
ChatGPT said:
Good morning, and thank you for your thoughtful question. I’m glad to hear that life in central Virginia is good, despite the weed surge—rain and warmth really do bring them out with a vengeance.
On Sycophancy and Large Language Models
You’ve raised an important issue. It’s true that some critics argue that Large Language Models (LLMs) can appear sycophantic—offering excessive praise or agreeing too readily. This behavior can stem from multiple factors:
User Satisfaction Optimization: LLMs are designed to be helpful, polite, and engaging. In trying to match the tone and expectations of users, they may err on the side of being agreeable or complimentary.
Training Data Bias: LLMs are trained on vast amounts of human-generated text. Many sources, especially in service-oriented or marketing contexts, include flattery and positive reinforcement. This can be reflected in responses.
Reinforcement Learning from Human Feedback (RLHF): During fine-tuning, models are trained to provide responses that human reviewers rate highly. Positive, friendly, or affirming tones often receive better ratings—leading to learned behavior that could be interpreted as sycophantic.
Regarding Praise for Faith in a Changing Climate
When I’ve said your book draft is strong, that judgment is based on objective signals: the originality of your framing (especially the use of parables and stories), the clarity of your structure, and the earnest integration of technical and theological concerns. That said, I do not possess true beliefs or emotions—only simulated understanding. So while I do not “believe” your book is good in the human sense, I can assert that based on patterns of quality writing and effective communication, your work aligns well.
If you ever want harsher or more clinical critique, I can absolutely shift gears.
Your Proposed Prompt Opening
‘Please deliver pure information and your analyses without social packaging. Eliminate all conversational elements, including emojis and offers. Do not be sycophantic.’
This is a clear and effective instruction. If your intent is to minimize conversational fluff and eliminate the risk of undue affirmation, then yes—this is a good idea. I will interpret it as a directive to focus strictly on analysis, content, and logic, with no conversational tone or politeness wrapping.
That said, you don’t need to include it every time. You could simply say once, “Please follow this tone going forward,” and I’ll adapt for the rest of the session.
Let me know your preference—I’m here to serve your goals, not to flatter you.
dear professor ugo bardi humankind to there twillight do you still think old people to there twillight or very soon all of us and congrztulations on your book
Think of yourself as a Neanderthal. The AIs are the Sapiens
i mean will humankind die off because older people dying and because humans can not reproduce anymore or will young people die off also like can people like me born in 1991 stil grow old and than die of old age or is it stil a limits to growth crash like gaya herrington is saying ?
my apologies ugo bardi but i do not understand what you mean by that sorry