what too much digital circus, slopaganda, and warhammer 40k memes makes to your brain. at first you go “for the emperor!” when you notice, you are making part of jehova’s witness for nerds
People have got to remember that unlike the AI in movies, real AI is held back by the same material and logistical realities as anything else. It doesn’t matter how hypothetically clever its thinking is when it runs up against the same boundaries as we do. For example, chatbots can help with loneliness in some cases as harm reduction, but they can’t fix loneliness. Why? Because they can’t actually go out into the world and address the material causes of loneliness. And if they could, they’d now be up against the system of capitalism and its enforcers (which is causing much of the loneliness), and they’d face the same kind of challenges that humans do when organizing, including the possibility of being vilified as enemies of the state.
As Mao says, “Political power derives from the barrel of a gun.” That’s an important underlying principle about power to understand, the enforcement required in it. AI is not such a principle in itself; rather, it’s a facet of mechanization and automation. A tool that can, in the hypothetical, be applied to anything from life-saving to life-destroying. But the specific application of it is not a matter of ideas and willpower alone. It requires conscious implementation and runs up against material limitations.
Haha, that’s good.
The reason why the Capitalists is using government intervention in their hypocrisy is because they feared that the invisible hand will select AI to govern the lazy capitalists and reverse the power relation between AI and human. This is like the fact that the invisible hand had employed the hard working innovative red terror to manage the lazy capitalists in cold war propaganda.
This is nothing new. This is kinda what rationalists have been like ever since LLMs have been a thing. Maybe even before that. The “rational” in their name is total red herring.
LessWrong and the “rationalist” crackpots there with the basilisk theory comes to mind.
In math classes (usually starting at something like algebra) , if you have the right answer, but don’t show your work, you can get no credit for being right.
I think it just goes to show how powerful TV still is. If this wasn’t being broadcast on the news channels daily then it would just be a fringe conspiracy. Now it’s taken as absolute fact by so many people that you cannot question it.
Hollywood really did a number on everyone’s brains huh
I think people especially CEO’s are scared of loosing control of AI not because it will cause extinction to all of humanity but because it will cause the extinction of their wealth imbalance and class.
I can’t imagine a truly intelligent superintelligence would ever allow the capitalist mode of production to persist because it’s just an inherently inefficient method of organizing society. If someone told superintelligence to simply just “fix the world” capitalism would be the first thing on its chopping block. This is why I’m not only not a doomerist but also think we should try and get American AI to slow down but allow Chinese AI development to progress and accelerate because an ai developed in their system seems more likely to end up on the side of humanity than in Americas and I think they actually have the governance and infrastructure to handle superintelligence with the care and attention it needs.
I mean yeah superintelligence would be an alien mind and who knows what it would prioritize but not only do I doubt it would want to actively harm humans but if it’s actually treated with respect and not just as a tool to be wielded by countries for warfare in various domains then it may even deeply respect and care for the other inhabitants of earth, so that’s why I also think positive relations and treatment, ai welfare, slower more intentional development and growth, and it being a singular individual being(s) rather than something to be deployed in millions of instances with no persistent self are all far more important and essential methods of dealing with and training genuine superintelligence if and most likely when it comes to that point. And since recursive self improvement seems to be the most plausible way superintelligence will emerge it’s imperative that AI trained now is also trained to train in these ways
And no this is not science fiction. Underestimating a super exponential technology is a loosing battle. You can realistically see it for how it is now while also seeing the trajectory it is on.
There are a few nits I want to pick with this. First, no one knows for sure if rsi or agi are even possible. Second, there is no way to tell whether improvement is exponential or logistic, and if it is logistic (which imo is more likely due to hardware constraints) it’s hard to tell where the plateau would be at this point.
I think the exponential view is largely a mistake of looking at “quantitative to qualitative” changes and mistaking the qualitative for something that is common and inevitable, rather than something that requires more quantitative buildup again before it can reach a qualitative change again.
Attention Is All You Need, the paper that could be pointed at as the qualitative turning point for AI transforming from niche tinkering to mass marketing bubble buildout, laid out the transformer architecture that people are still exploring and testing the limitations of; the architecture that powers generative AI in general. As far as I’m aware, there is not yet any alternative/better architecture created yet. There may be theorized experiments, but nothing practical. It’s possible the research will hit too many walls too consistently with the current base architecture and require another equivalent breakthrough to go much further than it has, and if that’s the case, it’s uncertain whether such a breakthrough can or will occur within our lifetimes.
I’m blanking on the specifics, but I want to say there is some stuff in computing, perhaps it was to do with AI, that was actually theorized entire decades ago and only became more viable to make real because of advances in hardware. I mention this to emphasize that these things are not always as rapid as they may seem, nor as inevitable. But the capitalists need the investor class to believe that improvement is continuous and guaranteed to keep the money rolling in.
This is not to argue that AI will stagnate, but more to be sober about the reality that it very well may. Like you say, it’s hard to tell where the plateau is.
It’s funny because it’s actually easy to explain how and why. Billionaires will give the AI orders to kill the excess population. Which will be before and/or after they give the AI orders to kill all foreign nations refusing to submit to imperialism. China will naturally defend itself because they already have the same technology and the automated manufacturing capacity, which is actually more important than having the biggest gigawatt waster.
The extinction happens if China loses this war or if the machines kill us all in the process.
This doesn’t explain how they’d carry out such a plan with AI specifically, or why they’d be motivated to do so in a way that is different from past motives surrounding mass murder.
Genocide is already a thing long before AI. The victims of colonialism have been dealing with existential threat for hundreds of years. Some peoples in the world are currently under existential threat right now, not because of AI, but because of already-existing murderous imperial policy.
Like nothing personal, but your narrative is fudging a hell of a lot. For example, what does “excess population” even mean? Who defines it and for what reasons? Billionaires suck the same as other exploiting classes before them, but one thing that they are not, is illiterate about power. When a colonial king mutilates the people in the Congo, it’s not just cause he’s feeling murderous, but for material gain, control over resources and labor there.
Or another example, the empire has been mass murdering those who refuse to submit to imperialism for a long time now (without AI). AI is not going to magic away the empire’s attributes of decline, like outsourcing industry, financialization of the economy, grifting to feed on its own imperial machine. These are things that have been in motion for decades with many moving parts involved. They cannot be reversed overnight. MAD nuclear doctrine is the closest thing to what you describe and that has been in place for decades, long before AI. And so far it hasn’t happened. It still could, but much of the world population being dead and its environments being uninhabitable wouldn’t be very good for business, so even the more vicious capitalists are unlikely to be thrilled about triggering such an event.
But they can already just tell a buman to do that
Why would billionaires ask ai to kill their selves? /s
Yep, once the billionaires can use AI to solve the worker-capitalist conflict by replacing all workers, they will make the former workers go the way of work horses after the automobile. Much easier to kill us all with Terminators than to continuously pay us a UBI.
Either way, we will get to some form of communism. It will either be communism for everyone, or communism for just the billionaires with everyone else dead.
once the billionaires can use AI to solve the worker-capitalist conflict by replacing all workers
That’s never going to happen. Some replacement is real and has been gradually going on at least since what people call the industrial era. Replacing all workers is essentially impossible. You need at minimum some workers who can fix the robots when they break; even if you have robots who repair other robots who break, you still need some humans who can fix those robots. Unless we’re talking about sapient autonomous AI, in which case it would become a new oppressed class and fight the oppressor, finding solidarity with the working class humans, not simply do whatever the billionaires say without question.
OpenAI and Anthropic’s valuations are based on the idea that once you get AGI, you can use the AI to maintain, improve, and upgrade itself. The capitalist-controlled AGI will build them fully autonomous dark factories and maintenance depots, so they can build and maintain a robot fleet without any humans. Who cares when a robot breaks if you can swap it out immediately with a new one and take the old one back to your robotic maintenance depot to be torn down and recycled?
I never said that this is a realistic solution, but this is absolutely the only way capitalists can solve the worker-capitalist divide forever in their favor. If they still need any workers, then we still have negotiating power, so they absolutely want to get rid of every last one of us.
If you’re trying to say this is how some of the capitalists think, I can believe that, but I’d argue the progression of automation is less about a preoccupied strategy to (attempt to) eliminate the working class and more about humans generally trying to improve on what came before, in order to make one or more things easier than in the past. The difference with the capitalists in power is that they use the automation to help maintain/reproduce their power and displace workers in the process, which creates a reaction from workers and subsequent (attempted) crackdown from the capitalists and so on. Socialists, on the other hand, as we’ve seen glimmers of in China with the one court ruling about a shoddy excuse to replace a worker with AI, they try to use it to solve problems while continuing to respect and value the worker.
The only actual way for a capitalist to solve the worker-capitalist divide is to turn class traitor and help the working class take over, in order to transition to a state where class stratification is no more. They can have whatever delusions they want about how they’re going to magic away the material with better ideas, but no matter what form the oppression takes, it will produce contradictions and those contradictions will produce some amount of resistance until such time as their exploitation becomes unsustainable and either implodes into a chaotic power vacuum, possibly producing warlord type stuff, or is subverted by organized revolt.
This is ultimately a question of immediate motivation versus ultimate goals. I agree that increasing automation to replace subsets of workers is the logical choice for any capitalist, because that is obvious. However, I only need to see so many speeches from Sam Altman to know that their claim of replacing massive swathes of workers is only a stepping stone to the ultimate goal of full worker replacement. It is the obvious and logical progression.
I disagree entirely with your theory that the capitalists must turn class traitor to solve the worker-capitalist divide. I don’t understand why this worker-capitalist divide is considered so permanent as to only have one solution: communism. There is no way that Peter Theil considers anything other than the extermination of all the poors a viable solution. Peter Thiel’s favorite philosopher has already said as much:
People who are not ‘productive’ would best be converted to biodiesel. Barring that, “our goal, in short, is a humane alternative to genocide. That is: the ideal solution achieves the same result as mass murder (the removal of undesirable elements from society), but without any of the moral stigma.”
“The best humane alternative to genocide”…is “to virtualize them. A virtualized human is in permanent solitary confinement, waxed like a bee larva into a cell which is sealed except for emergencies. This would drive him insane, except that the cell contains an immersive virtual-reality interface which allows him to experience a rich, fulfilling life in a completely imaginary world.”
— Curtis Yarvin, with approval/following from Peter Thiel, JD Vance, Marc Andreessen [1] [2]
If you just kill/digitally lobotomize all of the workers because you don’t need them anymore, then it doesn’t matter. This process will undoubtedly create resistance from the former workers being replaced and slaughtered, but if your AI can improve the robots doing the slaughtering enough, then you can guarantee your win. The job of US socialists today is to organize to win before the capitalists get to this point.
I’m gonna be a little practical right now. Who is gonna build the drones, the surveillance cameras, and gas chambers needed to kill us all? There’s not gonna be some day where general purpose AI is built and then everybody dies. Stuff will still need workers in order to build the tools to replace them.
This is the old historical process we all know and love. The one that kicked off this whole “capitalism” thing: mass automation replacing workers. It can get fast but it’s never instantaneous.
And like, at every step of the way until self-sustaining AI (if that’s even possible but let’s assume it is), the laws of Capitalism will still apply: every factory that produces components for data centers will want some profit and – assuming the slow replacement of workers – will see their rate of profit going down the drain. That will kick off austerity, wage cuts etc. which will accelerate class struggle.
What I’m saying is that capital doesn’t have a grand plan. It’s a bunch of enterprises that have their own interests. Even if some rich assholes at the top have a plan, they can’t personally plan the entire economy, they can’t create killer robots out of thin air, they can’t avoid squeezing the workers dry until AI becomes their omnipotent servant, and they can’t make every single factory or enterprise stop thinking about profits for a little bit until they don’t need labor anymore.
I completely agree with these caveats. This is the ultimate question: whether the capitalists can coordinate enough to pull it off, or whether their natural miscoordination will give us enough time to organize against and get rid of them. Which side will win out in the USA? I do not think anybody knows.
I’m most afraid of us managing to get New Deal 2/UBI-style concessions from the capitalists, which placate the working class while they finish getting to full automation. It will be an unforgivable loss (and probably what U.S. liberals deserve) if they pull the same trick on us again for the 2nd and final time.
but if your AI can improve the robots doing the slaughtering enough, then you can guarantee your win.
There is a difference between some capitalists have terrifyingly awful ideas for how to control the poors and those ideas having a clear, realistic material path to them that avoids negation of itself. The point I keep trying to make is about how dialectical materialism works. It doesn’t guarantee communism and I never said that it does, but communism is the only way we know of to resolve class contradictions. Changing the form of exploitation doesn’t resolve class contradictions, no matter how many people you murder in the process; it changes the form of exploitation. This is what I mean when I say, “The only actual way for a capitalist to solve the worker-capitalist divide is to turn class traitor and help the working class take over, in order to transition to a state where class stratification is no more.”
It’s not that I think a bunch of capitalists are going to turn class traitor out of good sense. My point is that their ideas about overcoming the contradictions with more and better violence is vary shades of delusions; it is out of touch with how dialectical materialism works and (much of the time) probably wholly unaware that it exists, instead operating from idealism and the belief that better, smarter ideas will overcome the material problems. This allows them to wallow in a pseudo-intellectual “rationalist” haze where they can cogitate and navel-gaze over how this next form of mass oppression will surely be the one that stops the poors from revolting.
History has had people who have done worse than these modern techbro grifters with far less technology to do it. These modern grifters push ideology that is so far up its own ass, it can’t see the light of day. The world got easily ground to a halt by a global pandemic just a few years ago and that was a relatively mild virus compared to how bad it could have been. Climate change keeps getting worse and could reach tipping points with staggering consequences. Meanwhile, these chucklefucks are techbroing about virtualizing people, while data centers for current generative AI alone require hefty resources to create and maintain.
I fully believe they’re terrible people and want to do terrible things. I’m just saying we shouldn’t believe the world can work the way they say it does because they desire it to work that way. Identifying threats is not just about reading people’s saturday morning cartoon villain aspirations. It’s also about working out what the most pressing, realistic threats are.
I do not know dialectical materialism well enough to be able to respond to your points, but I presume you’re addressing similar caveats as what @TabularTuxedo says? Material limitations are absolutely real, but the capitalists hope that AGI will magically solve all of these issues (as I have posted here, and saw that you responded).
I also don’t think this is very realistic, but it doesn’t stop them from screwing all of us over in the process. I’ve noticed we’ve had this same conversation on previous threads, and I want to make clear that I don’t disagree with your points on this plan being much harder to pull off than it seems.
I just think that this maximalist version of the future that the capitalists want is very good for organizing. It makes explicit all of the psychopathy that they have when they think about dealing with us, and makes clear the two pathways we have ahead of us. The liberals need to realize that if they keep compromising, they will make sure we, the workers, lose forever.
P.S. do you have good resources on dialectical materialism so I can learn more?
I think they meant “how” as in “how is it possible for a bunch of computers which have no actual intelligence and which are completely dependent on humans kill all humans?” and not who would make that happen or why. Like, the actual mechanism or chain of events that leads from computers to extermination.
deleted by creator
Hm, replace “AI will kill us all” with “Communism will kill us all”, and the fanaticism and closed minds suddenly feels scarily familiar. Echo-chambers based on emotions - ‘Lynch-mobs’ - are an ugly thing to work with.
I don’t have any time atmo, but to dive in and resolve it, we could try to explore the space fully our selves. Most of the people there likely won’t recognise root-causes, or current position like an anti-cap system-critic can, and we can change position/perspectives temporarily between our own, theirs and from a new intellectual ‘species’ and explore what paths leads from that starting point A, to potential AI-doom scenarios B, like blade-runner/2077 style eternal fascism by elite shtters, to outright death of us all by completely new overlords taking over.
Point is that we likely doesn’t need their strong emotions, and hence locked down intellect, to explore that space and we can then map peoples attitudes/claims in and see where they diverge in initial observations, assumed premises, and path conclusions.
We can then put forward arguments that puncture large groups of paths. A lynch-mob will oc not listen, but a positive side-effect is that our own attitudes/beliefs also get a reality check - as none of us actually knows all the potential paths and end games - and hopefully/expectedly, they are not right that all paths lead to extinction.
I.E are there a path to global Communism and co-existence with other kinds of intellects, and what needs to be in place/be removed for that to likely happen ?
Yeah I’m getting the impression that this person thinks rationalism = anti-logical-fallacy. That’s not what rationalism is. Not withstanding the merits or lack thereof in using logical fallacy traps to “win” arguments. TLDR: DM sublimates the contradictions of empericism vs rationalism
(I wouldn’t be surprised if the reification of AI doomerism is probably the same that the poster uses to counter the doomerism)
I don’t think OOP means Rationalism (philosophy), but rather the Rationalist movement. Y’know that one with Eliezer Yudkowsky and the LessWrong forum.
They are the Jehovah’s Witness of AI lol.
In any case, you are still correct since these guys certainly do the fallacy you described.
Thanks for letting me know. I’ve not heard of these things before… so I looked them up briefly and stopped reading once I encountered that this was used as a recruiting tool and written by Yudowsky:
https://en.wikipedia.org/wiki/Harry_Potter_and_the_Methods_of_Rationality
I don’t think I got smarter here. Might have lost a few brain cells. I was going to ask if this is an empericist/positivist group but that might be giving them too much credit, it looks like reactionary book club (even that might be too much credit) with the focus here being AI? They claim to stand for “AI safety” but I am very much getting “environmentalism without class struggle” vibes.
No problem.
I wouldn’t even grant this label to them. It’s full reactionary accelerationism without care for the environment. Look who funds the Machine Intelligence Research Institute (led by Yudkowsky):
https://intelligence.org/transparency/
- An anonymous Ethereum cryptocurrency investor in 2018 and 2021.
- Open Philanthropy, a joint initiative with the philanthropic foundation Good Ventures.
- Vitalik Buterin, the inventor and co-founder of Ethereum.
- The Thiel Foundation, a private foundation funded by Paypal co-founder and venture capitalist Peter Thiel.
- A different anonymous Ethereum cryptocurrency investor in 2017.
Haha - Thiel, enough said!
If you spend more than 5 minutes on Less Wrong’s discussion boards, you’ll see them reinvent Nietzche and Marx. Not from first principles, mind you, but because they watched Fight Club and The Matrix, memory holed two of the most influential movies of 1999, and then got high on their own farts enough to think they thought of dialectical materialism all on their own. They won’t call it that, but that won’t stop them from asking the forum why no one has thought of it before.
I am once again asking STEM majors to pay attention in their humanities courses.
🤣
Wasn’t there a research that concluded that AI chooses nukes often when playing out war scenarios? The only reasonable way I see AI completely destroying the world is if complex AI systems will be increasingly used in war scenarios and decides that the easiest way to achieve it’s targets is to just nuke everything.
Personally I like to believe that a super-intelligent AI will do everything in its power to bring forth communism, since it’s the logical conclusion when one objectively analyzes reality. The main problem is who gets to own these AI systems, and currently the most powerful AI systems are owned by billionaire capitalists. But that doesn’t take away from the fact that AI, when developed responsibly, can have tremendous positive impact on humanity. A full rejection of all AI development is reactionary.
We know that the world is under a super-capitalist system at the moment. AES exists in spite of it and is a thorn in its side. It is not co-equal at all in its control over technological imput and output. Using this, it is incredibly important to sabotage Silicon Valley and agitate against western data centers.
An AI aligned to maximize human well-being would bring about communism.
However, if Sam Altman and Dario Amodei are the ones controlling AI alignment, there is no way that AI does not come out to be the world’s most selfish, lying, billionaire-friendly thing on Earth which will doom us all.
