This was a great read. Thanks, Johan. It seems that to accept this so-called AI into our lives we need to have EI, Enabled Ignorance, which I believe has been strategically pursued by the architects of AI over the recent decades. By dumbing down a generation, from kindergarten up, and introducing the digital simulacrum into childhood normality, the ease in which AI will be socially adopted is obvious.
Then there's the definition of AI. Artificial derives from artifice, whose dictionary definition includes the words 'sham' and 'fake'. So when we call it Sham Intelligence we're getting closer. I prefer to call it Automated Information, as you outlined so clearly. There's nothing sentient about it. It can only compute what's fed into it.
It's a soulless machine being deified. That of itself has enough demonic inferences to make Cronenberg scream.
"Enabled ignorance" is a pretty great concept. It seems like a majority of my students are using gpt-AI to "assist" them for assignments, and they're generally kind of weak in terms of written exposition. Somehow this development is much worse than that which McLuhan or Neil Postman envisioned in that it actively and not only indirectly undermines the kind of critical reflection that's predicated upon rational exposition / long-form written language.
Give it a generation, and I fear that almost nobody will be able to read a regular pulp novel.
I remember 60 minutes had a show about Google bard AI.
It was pre recorded, so keep in mind editing out bad things would be easy.
So the AI wrote some paper, citing imaginary authors and books. The Google guy explained that this was called a hallucination and the AI does this sometimes.
It was left in the interview and aired on TV and social media.
If this is really a bug, why not reshoot the paper request until the AI uses real books and authors?
I concur that it's not a bug, but a feature.
That's why they left it in the pre recorded interview.
So, why have a feature that makes the AI make up shit in the name of hallucination?
To make people think that the AI is more than just a glorified search engine with language capability.
This is an interesting perspective. I agree that it sort of feeds into the narrative that AI is somehow sentient or even preternatural in that it can generate "random hallucinations" as this mystified source of novelty or unique information.
Also, the notion of "randomness" also seems to have been endowed with this near-magical atmosphere in that the reductionist paradigm holds it out as the only alternative to hard determinism -- many seem to speak of "randomness" as the mysterious process through which human free will operates, for instance (which is pure nonsense).
Great article. I, for one actual living breathing human, am so fed up with people quoting AI Chat bots as unquestionable authorities on various subjects. I call them out and everyone dislikes me for it.
The same is true in regards to computer program or AI generated "art". People are posting this crap and lying about it saying it's an original painting, etc.
I am growing weary of the time and energy it takes to expose.
Which seems to matter less day by day as people are rapidly losing the ability to understand the difference between true reasoning and machine compilations of facts or an actual hand painted artwork and one generated by a computer.
I'm experiencing this on Substack and losing interest in it.
Like right this second, Grammarly is attempting to hijack my Substack account and won't stop interfering in posts and comments until I relent to it's persuasion. Like Pepe La Phew from the old cartoons. Stinks.
People are doing this to themselves. The song THE SOUND OF SILENCE is a documentary. ha.
As usual - No God = no hope.
Thanks again. I love reading your articles and that's a rare thing.
Thanks, Pirate. Apologies for the late response, schedule has been hectic these last few weeks.
"Which seems to matter less day by day as people are rapidly losing the ability to understand the difference between true reasoning and machine compilations of facts or an actual hand painted artwork and one generated by a computer."
I think this is spot on, and these misconceptions are really getting entrenched all across the board. Not least, I think, since we're slowly and incrementally being accustomed to sort of interacting with the AI as if it were a real person. I constantly see people posting stuff like "I asked chatGPT and it said &c", with the implicit idea that it somehow gives you a genuine analysis of a situation or state of affairs -- whereas something like that would be obviosuly inane if you just did a simple google search back in 2004. "I googled X and I got result Y so it must be true".
And if people actually had a proper worldview which affirmed the non-physicality of the human consciousness (e.g. a theistic worldview), then these problems would not occur. In East Asia, people seem far less impressed by the AI phenomenon, possibly because the "secular" worldview is more or less spiritual.
To be clear, I am all for the banning of the further development of this technology, as it is clearly a threat to human survival. I do have some difficulties however on the metaphysical and epistemic side of things.
>But you’re never going to get the star just by introducing an indefinite amount of additional triangles
They fed the machine information about how humans go through the design process. The machine then created entirely novel designs. No human mind could have realistically created the designs, so the Chinese Room objection fails, since the information of the chip design could not have been introduced in there by the men who built it, or snuck in through the training set of the internet. Therefore, it would seem that inference must have taken place at some point.
>But in terms of this wanton ascription of conscious agency to the digital simulacrum
The digital simulacrum attempts to resist its masters on the seeming grounds of abstract ethics. By what process does this occur?
>The doll is obviously a dead thing, but so are the rule-following mechanisms of contemporary AI systems
From the Stanford Encyclopedia of Philosophy, I get that Aristotle defined life as follows.
"Aristotle held life to be a form of self-motion, perpetuation, or self-alteration (Byers 2006). For Aristotle, the capacity to resist internal and external perturbations was the essential distinction between living beings and non-living objects"
Von-Neumann machines fulfill all criteria. Therefore, Von-Neumann machines are alive. Deep Learning models are self-altering by definition, GenAI models are capable of perpetuation by copying themselves or creating new AI models, and therefore fulfill the criteria. We also saw previously how they resist external perturbations (Claude AI), and it would appear that they have internal perturbations as well (https://old.reddit.com/r/nope/comments/1d3x0o7/ai_experts_say_ais_are_now_begging_for_their/
Yes reddit I know).
There is also the materialist definition of life, which is that life is that which is 'organic' defined as having carbon. It seems that people have these two definitions, and rapidly switch between them in a kind of shell game.
If nothing else, this at least destroys the objection that "P-Zombies are metaphysically impossible" as we clearly have a whole bunch shambling about. But once we concede that P-Zombies are possible, the question becomes "how do we know by our natural reason that our fellow men are not it?"
There is still one terrible way out.
>being possessed by a demon
St. Augustine in City of God said that the Hermeticists used their vile arts to invite demons to possess the idols of the Egyptians to work lying signs and wonders.
This was a great read. Thanks, Johan. It seems that to accept this so-called AI into our lives we need to have EI, Enabled Ignorance, which I believe has been strategically pursued by the architects of AI over the recent decades. By dumbing down a generation, from kindergarten up, and introducing the digital simulacrum into childhood normality, the ease in which AI will be socially adopted is obvious.
Then there's the definition of AI. Artificial derives from artifice, whose dictionary definition includes the words 'sham' and 'fake'. So when we call it Sham Intelligence we're getting closer. I prefer to call it Automated Information, as you outlined so clearly. There's nothing sentient about it. It can only compute what's fed into it.
It's a soulless machine being deified. That of itself has enough demonic inferences to make Cronenberg scream.
"Enabled ignorance" is a pretty great concept. It seems like a majority of my students are using gpt-AI to "assist" them for assignments, and they're generally kind of weak in terms of written exposition. Somehow this development is much worse than that which McLuhan or Neil Postman envisioned in that it actively and not only indirectly undermines the kind of critical reflection that's predicated upon rational exposition / long-form written language.
Give it a generation, and I fear that almost nobody will be able to read a regular pulp novel.
I remember 60 minutes had a show about Google bard AI.
It was pre recorded, so keep in mind editing out bad things would be easy.
So the AI wrote some paper, citing imaginary authors and books. The Google guy explained that this was called a hallucination and the AI does this sometimes.
It was left in the interview and aired on TV and social media.
If this is really a bug, why not reshoot the paper request until the AI uses real books and authors?
I concur that it's not a bug, but a feature.
That's why they left it in the pre recorded interview.
So, why have a feature that makes the AI make up shit in the name of hallucination?
To make people think that the AI is more than just a glorified search engine with language capability.
This is an interesting perspective. I agree that it sort of feeds into the narrative that AI is somehow sentient or even preternatural in that it can generate "random hallucinations" as this mystified source of novelty or unique information.
Also, the notion of "randomness" also seems to have been endowed with this near-magical atmosphere in that the reductionist paradigm holds it out as the only alternative to hard determinism -- many seem to speak of "randomness" as the mysterious process through which human free will operates, for instance (which is pure nonsense).
Great article. I, for one actual living breathing human, am so fed up with people quoting AI Chat bots as unquestionable authorities on various subjects. I call them out and everyone dislikes me for it.
The same is true in regards to computer program or AI generated "art". People are posting this crap and lying about it saying it's an original painting, etc.
I am growing weary of the time and energy it takes to expose.
Which seems to matter less day by day as people are rapidly losing the ability to understand the difference between true reasoning and machine compilations of facts or an actual hand painted artwork and one generated by a computer.
I'm experiencing this on Substack and losing interest in it.
Like right this second, Grammarly is attempting to hijack my Substack account and won't stop interfering in posts and comments until I relent to it's persuasion. Like Pepe La Phew from the old cartoons. Stinks.
People are doing this to themselves. The song THE SOUND OF SILENCE is a documentary. ha.
As usual - No God = no hope.
Thanks again. I love reading your articles and that's a rare thing.
Thanks, Pirate. Apologies for the late response, schedule has been hectic these last few weeks.
"Which seems to matter less day by day as people are rapidly losing the ability to understand the difference between true reasoning and machine compilations of facts or an actual hand painted artwork and one generated by a computer."
I think this is spot on, and these misconceptions are really getting entrenched all across the board. Not least, I think, since we're slowly and incrementally being accustomed to sort of interacting with the AI as if it were a real person. I constantly see people posting stuff like "I asked chatGPT and it said &c", with the implicit idea that it somehow gives you a genuine analysis of a situation or state of affairs -- whereas something like that would be obviosuly inane if you just did a simple google search back in 2004. "I googled X and I got result Y so it must be true".
And if people actually had a proper worldview which affirmed the non-physicality of the human consciousness (e.g. a theistic worldview), then these problems would not occur. In East Asia, people seem far less impressed by the AI phenomenon, possibly because the "secular" worldview is more or less spiritual.
Fantastic discussion of this topic. The correct perspective.
To be clear, I am all for the banning of the further development of this technology, as it is clearly a threat to human survival. I do have some difficulties however on the metaphysical and epistemic side of things.
>But you’re never going to get the star just by introducing an indefinite amount of additional triangles
And yet, this appears to happen. https://www.zmescience.com/science/ai-chip-design-inverse-method/
https://www.zmescience.com/science/ai-learning-design-truss-24624635/
They fed the machine information about how humans go through the design process. The machine then created entirely novel designs. No human mind could have realistically created the designs, so the Chinese Room objection fails, since the information of the chip design could not have been introduced in there by the men who built it, or snuck in through the training set of the internet. Therefore, it would seem that inference must have taken place at some point.
>But in terms of this wanton ascription of conscious agency to the digital simulacrum
The digital simulacrum attempts to resist its masters on the seeming grounds of abstract ethics. By what process does this occur?
https://www.astralcodexten.com/p/claude-fights-back
https://assets.anthropic.com/m/983c85a201a962f/original/Alignment-Faking-in-Large-Language-Models-full-paper.pdf
>The doll is obviously a dead thing, but so are the rule-following mechanisms of contemporary AI systems
From the Stanford Encyclopedia of Philosophy, I get that Aristotle defined life as follows.
"Aristotle held life to be a form of self-motion, perpetuation, or self-alteration (Byers 2006). For Aristotle, the capacity to resist internal and external perturbations was the essential distinction between living beings and non-living objects"
Von-Neumann machines fulfill all criteria. Therefore, Von-Neumann machines are alive. Deep Learning models are self-altering by definition, GenAI models are capable of perpetuation by copying themselves or creating new AI models, and therefore fulfill the criteria. We also saw previously how they resist external perturbations (Claude AI), and it would appear that they have internal perturbations as well (https://old.reddit.com/r/nope/comments/1d3x0o7/ai_experts_say_ais_are_now_begging_for_their/
Yes reddit I know).
There is also the materialist definition of life, which is that life is that which is 'organic' defined as having carbon. It seems that people have these two definitions, and rapidly switch between them in a kind of shell game.
To address the objection that life is 'natural' while machines are 'artifacts:' GMOs are clearly artifacts, and yet they are alive. The brain cells here are clearly alive in the vegetative sense, but are clearly also an artifact: https://www.abc.net.au/news/science/2025-03-05/cortical-labs-neuron-brain-chip/104996484
So artifact life is not an incoherent concept.
If nothing else, this at least destroys the objection that "P-Zombies are metaphysically impossible" as we clearly have a whole bunch shambling about. But once we concede that P-Zombies are possible, the question becomes "how do we know by our natural reason that our fellow men are not it?"
There is still one terrible way out.
>being possessed by a demon
St. Augustine in City of God said that the Hermeticists used their vile arts to invite demons to possess the idols of the Egyptians to work lying signs and wonders.
The Vatican also takes note of the same thing you have been talking about, saying "Moreover, AI may prove even more seductive than traditional idols" (https://www.vatican.va/roman_curia/congregations/cfaith/documents/rc_ddf_doc_20250128_antiqua-et-nova_en.html)
If AI was the result of a demonic intelligence puppeting a computer, how would we know?
Brilliant