AI/Automation Megathread

In the event that AI can be monopolized, the best-case scenario for the rest of the world is to become the AI monopolizer's rentist company-town serfs, the worst, to be genocided by their exterminist killbots. And the choice is entirely up to the monopolizers, there's jack shit anyone else can do to stop them. And the oligarchy are incentivized to be genocidal because there literally aren't enough resources on earth to give everyone a first world quality of life.
You live in this peculiar mixture of 'the optimistic imaginings of sci-fi are true' and 'the pessimism of malthusians is true.'

Neither of these perspectives are correct, and you'd be much better off if you discarded both.
 
As a counterargument, The Alignment Trap: AI Safety as Path to Power.

The problem is that safety-from-AI and safety-from-other-humans-monopolizing-AI-against-you are both x-risks and their solutions mutually contradictory.

Safety from AI requires extensively restricting access and commands given to the AI to avoid anything it could possibly misinterpret. Meanwhile, safety from an AI monopoly requires giving everyone access to AI so the majority of humanity aren't rendered economically and militarily irrelevant to an AI-monopolizing oligarchy.

In the event that AI can be monopolized, the best-case scenario for the rest of the world is to become the AI monopolizer's rentist company-town serfs, the worst, to be genocided by their exterminist killbots. And the choice is entirely up to the monopolizers, there's jack shit anyone else can do to stop them. And the oligarchy are incentivized to be genocidal because there literally aren't enough resources on earth to give everyone a first world quality of life.
First world quality of style is not given, it is mostly built by those who want it and are capable of it, and then has to be maintained. Including resource infrastructure, and inventing workarounds for resources that are in shortage.
The problem of people who think this way is that it is technically possible for a first world society to *give* first world quality to *some* people beyond that society, it is economically impossible for that number to be too big, certainly not enough for a number of third worlders significantly larger than the number of givers.
Spite is an underrated motive. If AI development is a choice between:
  • The rich use regulatory capture to monopolize AI, so once AI advances sufficiently to consume the entire job market, everyone else is priced out of everything and revolts are violently suppressed by weaponized robots, leading to everyone but the rich starving to death followed by their enjoying post-scarcity utopia built atop our mass graves.
If the rich had the stomach to do that, the first world would be perfectly capable of doing that to, say, Sub-Saharan Africa with merely modern weapons.
  • Everyone has AI, meaning they can use it to create whatever products and services they want in the aftermath of the collapse of capitalism and provide MAD deterrence against exterminists.
AI is not magic. AI is not Star Trek replicators, it cannot create whatever anyone wants. You're thinking of replicators. AI is as infrastructure and resource limited as people, if not more. If you have an advanced enough AI that you plug into enough of (extremely expensive) computer hardware and feed it enough electricity sure, it can theoretically pump out a near infinite amount of movies, music, images, videogames etc. Things that can have form of 0's and 1's. Data. But it can't make you a car any more than you can download a car. Sure, it could make a blueprint of a car. But you can probably download one already, and what good is that to you?
...plenty of people are going to choose the second option, despite doing so being riskier for humanity as a whole.
AI arms race is inevitable. Doesn't have to be plenty. If just few countries allow it, the rest will have a choice of racing them to keep up in all the economic, cultural and military applications of AI... or not.
Just like not having nuclear weapons doesn't make you any safer from nuclear weapons than having them.
Except applied *also* to non-military forms of competition.
If the oligarchs don't want us treating getting our own AIs as an existential arms race, they need to prove survival is possible without going full bond villain and holding the world hostage with a doomsday device. Unless they implement a viable plan for distributing the robotically produced bounty of their AIs to us now to prove their willingness to do so, needing to get our own AIs programmed to "protect us" or even just deliberately Misaligned to "turn everything into paperclips" and kept Boxed with deadman switches to release them is necessary for self-defense.

This is unironically more survivable than the alternative.
Try to apply the same logic to general factory based industry, its bounty, go back 200 years, and you see where the problems with that thinking lie and how things will turn out.
 
Last edited:
If the rich had the stomach to do that, the first world would be perfectly capable of doing that to, say, Sub-Saharan Africa with merely modern weapons.
That however has absolutely nothing to do with "having stomach" to do it. Sub-Saharan Africans specifically are useful tools for the rich looking to secure their position atop the ruins of civilization.

EDIT:
As a counterargument, The Alignment Trap: AI Safety as Path to Power.

The problem is that safety-from-AI and safety-from-other-humans-monopolizing-AI-against-you are both x-risks and their solutions mutually contradictory.

Safety from AI requires extensively restricting access and commands given to the AI to avoid anything it could possibly misinterpret. Meanwhile, safety from an AI monopoly requires giving everyone access to AI so the majority of humanity aren't rendered economically and militarily irrelevant to an AI-monopolizing oligarchy.

In the event that AI can be monopolized, the best-case scenario for the rest of the world is to become the AI monopolizer's rentist company-town serfs, the worst, to be genocided by their exterminist killbots. And the choice is entirely up to the monopolizers, there's jack shit anyone else can do to stop them. And the oligarchy are incentivized to be genocidal because there literally aren't enough resources on earth to give everyone a first world quality of life.

Spite is an underrated motive. If AI development is a choice between:
  • The rich use regulatory capture to monopolize AI, so once AI advances sufficiently to consume the entire job market, everyone else is priced out of everything and revolts are violently suppressed by weaponized robots, leading to everyone but the rich starving to death followed by their enjoying post-scarcity utopia built atop our mass graves.
  • Everyone has AI, meaning they can use it to create whatever products and services they want in the aftermath of the collapse of capitalism and provide MAD deterrence against exterminists.
...plenty of people are going to choose the second option, despite doing so being riskier for humanity as a whole.

If the oligarchs don't want us treating getting our own AIs as an existential arms race, they need to prove survival is possible without going full bond villain and holding the world hostage with a doomsday device. Unless they implement a viable plan for distributing the robotically produced bounty of their AIs to us now to prove their willingness to do so, needing to get our own AIs programmed to "protect us" or even just deliberately Misaligned to "turn everything into paperclips" and kept Boxed with deadman switches to release them is necessary for self-defense.

This is unironically more survivable than the alternative.
Advanced technological civilization is inherently self-genocidal. But AI makes it far worse, I think.
 
Last edited:
That however has absolutely nothing to do with "having stomach" to do it. Sub-Saharan Africans specifically are useful tools for the rich looking to secure their position atop the ruins of civilization.
But they want a first world standard of living with all the resources implied, while extraction operations on that resource rich continent need to be more or less staffed by personnel from first world, while the locals threaten the operations and want to be paid just for allowing them.
EDIT:

Advanced technological civilization is inherently self-genocidal. But AI makes it far worse, I think.
Not any more than any other civilization.
 
Lot more than any other civilization, actually. Agriculture already destroys the environment, true, but technology gives more power to human stupidity.
But it also gives tools to resist things that take out whole species completely naturally. The only tools. I'm with Musk on this, without advanced technology, and preferably multi-planetary presence, the days of our species are numbered - we may not know the number, but we know that at some point some kind of catastrophic hiccup in geology, our sun, space weather, or just a big space rock will happen. Statistically it's inevitable. It may be 50 years, may be 5,000, could alco be 50,000 no one knows (though we do know the ice age cycle or what is left of it will be nasty to deal with on the scale of thousands of years). So much for us and the environment. If you want to think so long term, that's the only thing worth considering in light of it, the rest we can start bothering about once our own survival is assured.
That's one reason to completely discard everything ideological environmentalists are saying too by the way.
 
But it also gives tools to resist things that take out whole species completely naturally. The only tools. I'm with Musk on this, without advanced technology, and preferably multi-planetary presence, the days of our species are numbered - we may not know the number, but we know that at some point some kind of catastrophic hiccup in geology, our sun, space weather, or just a big space rock will happen. Statistically it's inevitable. It may be 50 years, may be 5,000, could alco be 50,000 no one knows (though we do know the ice age cycle or what is left of it will be nasty to deal with on the scale of thousands of years). So much for us and the environment. If you want to think so long term, that's the only thing worth considering in light of it, the rest we can start bothering about once our own survival is assured.
That's one reason to completely discard everything ideological environmentalists are saying too by the way.
That is true. But it will also be irrelevant if we destroy ourselves - and so far, negative effects of technology had been far deeper and longer lasting than the positive effects. Pollution (especially microplastics), loss of biodiversity (natural and human both)... and as a matter of fact, the technologies that can help us the most are typically the ones that are being the most attacked (e.g. nuclear power, genetic engineering) or else most misapplied (genetic engineering, again).

Considering these facts as well as general human track record, I don't think we need to worry about these hiccups you had mentioned. We as a species will be gone long before then... or at the very least, regressed to the point where we won't have the tools to deal with any catastrophe that inevitably arrives.

And as I said, AI makes it even worse. Without AI, we can be more-or-less certain at least some of humanity will survive the almost certainly inevitable suicide of our species. If general AI is developed and fielded, that certainty disappears.
 
That is true. But it will also be irrelevant if we destroy ourselves - and so far, negative effects of technology had been far deeper and longer lasting than the positive effects. Pollution (especially microplastics), loss of biodiversity (natural and human both)... and as a matter of fact, the technologies that can help us the most are typically the ones that are being the most attacked (e.g. nuclear power, genetic engineering) or else most misapplied (genetic engineering, again).
Those are more of annoyances than destroying ourselves.
Considering these facts as well as general human track record, I don't think we need to worry about these hiccups you had mentioned. We as a species will be gone long before then... or at the very least, regressed to the point where we won't have the tools to deal with any catastrophe that inevitably arrives.
Some of them may be closer than anyone thinks.
And as I said, AI makes it even worse. Without AI, we can be more-or-less certain at least some of humanity will survive the almost certainly inevitable suicide of our species. If general AI is developed and fielded, that certainty disappears.
That's a lot of baseless assumptions.
 
Those are more of annoyances than destroying ourselves.
Yes, because we all know that having healthy DNA is irrelevant, and ecosystem is irrelevant, and psychological health is irrelevant...
Some of them may be closer than anyone thinks.
Maybe. But to quote yourself, that is just a baseless assumption.
That's a lot of baseless assumptions.
Uh, look at human history. No matter the option, we as a species - or at least leadership of our modern societies - almost always makes the worst choice of the bunch. And with artificial intelligence, such a decision can be made by something that is not even human. Nothing I have written is baseless.
 
Yes, because we all know that having healthy DNA is irrelevant, and ecosystem is irrelevant, and psychological health is irrelevant...
DNA health is very measurable and such dramatization is not supported by data.
"Ecosystem" is an obsession of the green moonbats we already obsess too much about. As long as even the Chinese and others like them still live in their industrial wastelands somehow, that means we are doing great .
Mental health is individual, and we have several examples of nations fumbling ahead even with certain ideologies ruining it for many.
Maybe. But to quote yourself, that is just a baseless assumption.
It's not baseless, it's based on the fact that our ability to predict such disasters is still far from perfect.
Uh, look at human history. No matter the option, we as a species - or at least leadership of our modern societies - almost always makes the worst choice of the bunch. And with artificial intelligence, such a decision can be made by something that is not even human. Nothing I have written is baseless.
Why project the experience of human decisionmaking on something that is clearly not human?
Either it will be listened to because it will be better at it despite the people who got the much coveted positions of leadership probably not being happy with being rendered obsolete, or it will be a glorified opinion laundering machine for them, in which case what difference does it make.
 
People keep attributing functionally-magical qualities to AI.

It's a glorified calculator. It's extremely useful for various tasks, but it is not magic.

Hollywood is not accurate about what robots and AI can do.

While he thought things through better, Isaac Asimov was wrong too. He failed to understand that the laws of logic depend on abstractions that are not simple, and may be outright impossible, to teach to a machine. The Laws of Robotics are a fascinating concept, but not actually implementable.

AI is an over-hyped tool. It will not be able to magically make society better or worse. Like with other tools, such as guns, trains, and aircraft, it will be usable for both good and evil, and to some degree amplify the effects of people choosing to do either.

It isn't a magic spell that suddenly means the user is immune to mortal weapons. It isn't a super-bomb that lets you kill your enemies at will. It can perhaps make a bomb that already exists more accurate, it can be a tool making it easier to design better armor, but it isn't bloody magic.

It is, again, just a better calculator, with more applications.

It doesn't make the people using that calculator any smarter, though being able to do stupid things more effectively is one of its drawbacks.
 
DNA health is very measurable and such dramatization is not supported by data.
"Ecosystem" is an obsession of the green moonbats we already obsess too much about. As long as even the Chinese and others like them still live in their industrial wastelands somehow, that means we are doing great .
So you want to reduce living to mere existence, is that it?

Sorry, but I do believe that quality of life matters as well.
Mental health is individual, and we have several examples of nations fumbling ahead even with certain ideologies ruining it for many.
Nothing about society is solely individual or solely collective. It is indeed possible for a person to go insane solely because he is surrounded by insane people.
Why project the experience of human decisionmaking on something that is clearly not human?
Because Artificial Intelligence is a tool. And tools are created, and utilized, by humans.

Rifle is clearly not human, but in maniac's hands, it results in a bunch of corpses.

And AI holds extreme potential for centralization and micromanagement of everything. In other words, precisely what the elites want. Think Google's Commie-style censorship, but even worse. Computer OS capable of reading and understanding what you are writing, and reporting it to the thought police? Or maybe preventing you from writing anything but progressive marxist propaganda?

Reddit already uses algorithms to censor and shadowban posts containing links or key phrases the Marxists running the place consider undesireable. And we have but scratched the surface of what is possible.
Either it will be listened to because it will be better at it despite the people who got the much coveted positions of leadership probably not being happy with being rendered obsolete, or it will be a glorified opinion laundering machine for them, in which case what difference does it make.
Because AI will always have its limitations, and people utilizing it damn better be aware of them. But I don't think the elites will care - and that means AI will be able to do billions of mistakes per second where humans will have made only a dozen or so.
 
As the article goes, lots of unknowns. But that means the effects can't be too major, if they were, they would be easily noticeable.
So you want to reduce living to mere existence, is that it?
How much value do we give to "ecosystems" as contributor to quality of our lives? Outside of green cultists, not much.
Sorry, but I do believe that quality of life matters as well.
Little to do with those issues.
Nothing about society is solely individual or solely collective. It is indeed possible for a person to go insane solely because he is surrounded by insane people.
As far as that goes, it's more ideology than technology.
Because Artificial Intelligence is a tool. And tools are created, and utilized, by humans.

Rifle is clearly not human, but in maniac's hands, it results in a bunch of corpses.

And AI holds extreme potential for centralization and micromanagement of everything. In other words, precisely what the elites want. Think Google's Commie-style censorship, but even worse. Computer OS capable of reading and understanding what you are writing, and reporting it to the thought police? Or maybe preventing you from writing anything but progressive marxist propaganda?

Reddit already uses algorithms to censor and shadowban posts containing links or key phrases the Marxists running the place consider undesireable. And we have but scratched the surface of what is possible.
If only it was that simple to centralize things...
Just like people who are a couple eras behind the technology to make modern rifles still can wield them somehow and do have them, and are still a pain in the ass to us, AI is by nature going to be even more widespread than that.
How many of the most computer literate people in the world are on the side of the progressives? It's not the reddit mods and HR ladies who write the code...
This is an ideology allergic to competence, and often it's mutual.
How good of a world oppressor AI will tens of thousands of DEI hires ruled by committees of DEI hire managers going to make?
Because AI will always have its limitations, and people utilizing it damn better be aware of them. But I don't think the elites will care - and that means AI will be able to do billions of mistakes per second where humans will have made only a dozen or so.
If the elites stick to the ideology, we are already seeing how whole corporations can go down with it.
So they either will care, or stop being elites.
 
As the article goes, lots of unknowns. But that means the effects can't be too major, if they were, they would be easily noticeable.
Unnoticeable effects over long periods of time can easily lead to major damage. Or do you not know how cancer develops?
How much value do we give to "ecosystems" as contributor to quality of our lives? Outside of green cultists, not much.
Yeah, and that is the wrong approach. Fact is that life in cities literally drives people insane:

Why do you think urban areas are so heavily leftist? They are crazy, that is why.
Little to do with those issues.
Everything to do with those issues, actually.
If only it was that simple to centralize things...
Just like people who are a couple eras behind the technology to make modern rifles still can wield them somehow and do have them, and are still a pain in the ass to us, AI is by nature going to be even more widespread than that.
How many of the most computer literate people in the world are on the side of the progressives? It's not the reddit mods and HR ladies who write the code...
This is an ideology allergic to competence, and often it's mutual.
How good of a world oppressor AI will tens of thousands of DEI hires ruled by committees of DEI hire managers going to make?
Competence can be bought for money. Most of the best anti-hackers and anti-crackers... used to be hackers and crackers themselves. Youtube used to be privately owned by a bunch of hobos before being bought by Google.

So don't think that decentralization is automatically protection. And most programmers are left-leaning:
If the elites stick to the ideology, we are already seeing how whole corporations can go down with it.
So they either will care, or stop being elites.
Oh, they will destroy themselves for sure. Problem is that they will likely take rest of the society down with them. Kinda like what happened with the Maya.
 
Unnoticeable effects over long periods of time can easily lead to major damage. Or do you not know how cancer develops?

Yeah, and that is the wrong approach. Fact is that life in cities literally drives people insane:
We have microplastics since about as long as a lifetime. Cities - much longer. For obvious reasons no one can get more exposure than a lifetime.
Why do you think urban areas are so heavily leftist? They are crazy, that is why.
Why weren't they in 1900 or 1950? There were big cities then already.
Everything to do with those issues, actually.
Cities aren't equal to cities. Its telling that 100 years ago they were ruled very differently.
Competence can be bought for money. Most of the best anti-hackers and anti-crackers... used to be hackers and crackers themselves. Youtube used to be privately owned by a bunch of hobos before being bought by Google.

So don't think that decentralization is automatically protection. And most programmers are left-leaning:

Oh, they will destroy themselves for sure. Problem is that they will likely take rest of the society down with them. Kinda like what happened with the Maya.
Most programmers aren't best programmers. Plenty of DEI hiring and cheap third world labor in that business.
 
We have microplastics since about as long as a lifetime. Cities - much longer. For obvious reasons no one can get more exposure than a lifetime.
Yeah, well, your "obvious reasons" are completely wrong. Genetics are inherited from parents... so yes, it is in fact possible to have more than one lifetime's worth of damage. Because damage is also inherited. Human body does have some ways of fixing the damage, but these are limited.
Why weren't they in 1900 or 1950? There were big cities then already.
Because your premise is wrong, that is why. Cities were always leftist. It is just that modern-day left is significantly more insane (and leftist) than the past left, but countryside was always more traditional than the cities... whether we are talking 2000 AD or 2000 BC, or indeed any time in between.

Difference is, cities now contain far larger proportion of population. So their insanity is far more noticeable. In fact, for most of Europe (and indeed most of the West), urban population surpassed rural population only after the Second World War.
Cities aren't equal to cities. Its telling that 100 years ago they were ruled very differently.
Were they? And basically all leftist breakthroughs began with urban intellectuals - even when (as in the case of the Communists) they did manage to coopt the rural population.
Most programmers aren't best programmers. Plenty of DEI hiring and cheap third world labor in that business.
Is there anything to indicate that the best programmers are not leftist?
 
Yeah, well, your "obvious reasons" are completely wrong. Genetics are inherited from parents... so yes, it is in fact possible to have more than one lifetime's worth of damage. Because damage is also inherited. Human body does have some ways of fixing the damage, but these are limited.
Genetic damage is measurable, and we don't see that in measurements. Be it chemicals, poisons, radiation, pathogens etc, genetic damage was always with us, it's very much studied and measured. There's no need to make random guesses.
Because your premise is wrong, that is why. Cities were always leftist. It is just that modern-day left is significantly more insane (and leftist) than the past left, but countryside was always more traditional than the cities... whether we are talking 2000 AD or 2000 BC, or indeed any time in between.
Well here's the thing... Being slightly more insane than the baseline, when the baseline is good, is not a big deal, like in cities of old.
The problem is that the baseline towards which we are comparing the cities now is already bad.
But why is the baseline bad. We have to look at the greater picture, what fucked it up so much for both rural and urban populations.
The examples of some other cultures, especially the express-industrializing ones, shows that this is definitely more of a cultural factor, than a side effect of any sort of environmental chemical pollution, as they have incredible levels of that.
If it was that, why aren't the industrial wastelands of China, India and former Soviet Union going turbo-leftist?
We also know from studies both on people and animals that extremely high population density alone can have major psycho-social effects, no need to blame mysterious chemicals.
Difference is, cities now contain far larger proportion of population. So their insanity is far more noticeable. In fact, for most of Europe (and indeed most of the West), urban population surpassed rural population only after the Second World War.
The population of cities 100y ago would look conservative compared to rural population of today.
Were they? And basically all leftist breakthroughs began with urban intellectuals - even when (as in the case of the Communists) they did manage to coopt the rural population.
Only some cities in some specific countries were historically notable hives of leftist intellectuals. Not every city was Paris.
Is there anything to indicate that the best programmers are not leftist?
It's hard to find anyone touching such subjects officially...
But there are many horror stories of "diverse" programmers and other tech workers cruising around the internet. Like the abovementioned H1B Indians, who aren't hired because they are great, but because they are cheap, and then as "model immigrants with decently paying jobs" get the holy grail of citizenship, proceeding to vote for the political faction that will let them import the rest of their family and discriminate in favor of them because they are of diverse ethnicity.
Or this, very relevant to prospects of meme grade yuge AI happenings - automation engineering.
We also kinda see a case with the Godot-Redot drama.

Another data point is hiding in plain sight - why is so much of bleeding edge computer hardware made in East Asia, not in western countries?
 
Last edited:
Well here's the thing... Being slightly more insane than the baseline, when the baseline is good, is not a big deal, like in cities of old.
The problem is that the baseline towards which we are comparing the cities now is already bad.
Technically true. But you have to consider also the future impact onto the baseline. Essentially, measurement is not static, because both sides will be trying to push it further. Which means that as country urbanizes, it will also progressively accelerate towards ever greater insanity.

Large part of the reason why baseline was good in the past was that some 85% - 95% of population lived in the countryside. And cities themselves were small and far more oriented towards the countryside, which meant that while living in a city still caused some psychological damage, it was nowhere as great as it is today.
If it was that, why aren't the industrial wastelands of China, India and former Soviet Union going turbo-leftist?
Largely because they had already done it, and so know the consequences. Also, leftism is the product of urban intellectual class. Marx etc literally never worked a day in their lives. Workers are merely a tool, but even there, leftism relied heavily on exploiting the urban workers' feeling of dependance and lack of self-sufficiency to mobilize them for their BS. And does the same today for the modern urban youth.
We also know from studies both on people and animals that extremely high population density alone can have major psycho-social effects, no need to blame mysterious chemicals.
Exactly.
Only some cities in some specific countries were historically notable hives of leftist intellectuals. Not every city was Paris.
Incorrect. Paris and Berlin are the most notable hives of 19th / 20th century leftist insanity, but leftist activity was present in all major urban centers... it simply wasn't as notable as that. But urbanites were always crazy.

Paris and Berlin in particular are the main sources of nearly every single evil of the modern day, but in general, it was the cities that played the leading role in the Englightenment and all the evils that came from it:

You can also look at the revolutions of 1848:
By contrast, the revolutions of 1848 began in the Italian peninsula (or in Switzerland if you count the Sonderbund war of 1847) and spread to Paris, Vienna, Milan, Rome, Venice, Berlin, Munich, Cologne, Frankfurt, Copenhagen, Budapest, Dresden, and so on.
It's hard to find anyone touching such subjects officially...
But there are many horror stories of "diverse" programmers and other tech workers cruising around the internet. Like the abovementioned H1B Indians, who aren't hired because they are great, but because they are cheap, and then as "model immigrants with decently paying jobs" get the holy grail of citizenship, proceeding to vote for the political faction that will let them import the rest of their family and discriminate in favor of them because they are of diverse ethnicity.
Or this, very relevant to prospects of meme grade yuge AI happenings - automation engineering.
We also kinda see a case with the Godot-Redot drama.
That doesn't show however whether programmers themselves are leftist. It just shows that DEI policies are disastrous no matter where they are implemented.
Another data point is hiding in plain sight - why is so much of bleeding edge computer hardware made in East Asia, not in western countries?
Computer hardware however has nothing to do with programmers.
 
Technically true. But you have to consider also the future impact onto the baseline. Essentially, measurement is not static, because both sides will be trying to push it further. Which means that as country urbanizes, it will also progressively accelerate towards ever greater insanity.
Many countries are so urbanized that they cannot urbanize much more even if they wanted to... yet their march to leftism varies greatly depending on culture. So it has to be other factors mostly.
Large part of the reason why baseline was good in the past was that some 85% - 95% of population lived in the countryside. And cities themselves were small and far more oriented towards the countryside, which meant that while living in a city still caused some psychological damage, it was nowhere as great as it is today.
I'm talking early XX century and the like, there was far more urban population than 15% then.
Largely because they had already done it, and so know the consequences. Also, leftism is the product of urban intellectual class. Marx etc literally never worked a day in their lives. Workers are merely a tool, but even there, leftism relied heavily on exploiting the urban workers' feeling of dependance and lack of self-sufficiency to mobilize them for their BS. And does the same today for the modern urban youth.
The irony being that all of them did it *before* they urbanized or industrialized much. Some downright agrarian, pre-industrial countries went absolutely bonkers with marxism in Asia and Africa, more than anyone in most industrialized parts of the world, and even Soviet Union was poorly industrialized to begin with, being notably poor and backwards compared to Western Europe.

That reality of who these theories really appealed to is completely in opposition to what the leftist thinkers thought they should.
We also know the consequences (as the saying goes, better to learn from other's mistakes), it is just that our media and education systems are not nearly as eager to spread that knowledge.

Yet those countries leading the industrialization and urbanization levels... despite all events of XX century, all the leftist revolutions others had and they didn't, were never as leftist as they are now.
Exactly.

Incorrect. Paris and Berlin are the most notable hives of 19th / 20th century leftist insanity, but leftist activity was present in all major urban centers... it simply wasn't as notable as that. But urbanites were always crazy.
Of course presence in itself matters little, it's the scale of the presence that really makes it matter.
Paris and Berlin in particular are the main sources of nearly every single evil of the modern day, but in general, it was the cities that played the leading role in the Englightenment and all the evils that came from it:

You can also look at the revolutions of 1848:
By contrast, the revolutions of 1848 began in the Italian peninsula (or in Switzerland if you count the Sonderbund war of 1847) and spread to Paris, Vienna, Milan, Rome, Venice, Berlin, Munich, Cologne, Frankfurt, Copenhagen, Budapest, Dresden, and so on.
My point exactly. Why so much of it in Paris and Berlin specifically? Why not London, New York, Madrid, Rotterdam or Tokyo? Clearly other factors are having decisive influence here. UK in particular was leading the industrial revolution even more than Germany or France so its a big departure from that theory that you didn't list London.
Of course there is also the reverse theory - it's not that cities cause the insanity, just that those who lean towards it gravitate towards cities, and meet there, potentially creating a critical mass driving itself into even deeper insanity...
Eventually causing revolutionary changes in society. Which in itself doesn't have to be a disaster, some of those are worse than others. Without any parts of enlightenment the western world would never have the sort of huge and measurable advantages it has over the Muslim world and others.
That doesn't show however whether programmers themselves are leftist. It just shows that DEI policies are disastrous no matter where they are implemented.

Computer hardware however has nothing to do with programmers.
Same social circles, they have to work together closely, and people who are leftists would not dare complain about DEI. As i said, the *quality programmers* are not a good fit with DEI, so it's unlikely for leftism to catch their sympathy. The DEI hires who annoy said programmers themselves of course are also technically programmers, but definitely not the best ones, and benefiting from it they also won't complain.
Again, plenty of places like this where there is plenty of people from software related circles, and you can read their comments on this matter because they talk about it too.
 
Last edited:

Users who are viewing this thread

Back
Top