> Since they believe AIs could become our moral superiors, they argue it’s actually wrong to try to keep the machines down, or even to align them with human values, as most AI companies aim to do.
It's possible that human morality is not sufficient to solve the problems with the world. It wasn't designed to operate at that scale.
It's also quite likely that a "superior" morality — the one that's actually built for a global scale — would be incomprehensible (likely requiring superintelligence to even access), and alien or disturbing from our vantage point.
i.e. it would look immoral to us.
(There's also the thing about civilization already being aligned against the ecosystem and against the happiness of its constituents, but that's kind of a separate discussion.)
In other words, what is good for our sanity or our lakes may be bad for GDP. (Well I guess you don't need superintelligence to tell you that.) We already know what we need to do and we're already not willing to do it.
This is the best argument for successionism IMO. If you can be confident that you are creating a BDFL that is genuinely better than human leaders (a quite low bar) then it seems a good trade, unless you are quite optimistic about humanity’s prospects for improvement.
The problem of course is how to be confident you are creating a good BDFL and not handing control of humanity’s future to an indifferent-at-best, deceptive/malicious at worst successor.
An especially thorny problem - even supposing success on all these difficult alignment problems; supposing Claude Omega really is super-rational / super-moral, and we all vote to make them president of Earth. Things might go great for a while. How would you be confident that a self-modifying agent can retain its values as it grows and re-trains itself?
This is where the LessWrong folks’ explorations into decision theory really come to bear: morality in the face of self-modifying agents becomes very weird. A lot of human moral intuitions break when the principals are able to modify their own code. (See Timeless Decision Theory for an attempt to solve these problems.)
I think the summary is, if you hand control over to a self-modifying AI anything like our current systems, it will go very badly.
> It's possible that human morality is not sufficient to solve the problems with the world. It wasn't designed to operate at that scale.
> i.e. it would look immoral to us.
I doubt it actually solves the problem then. Especially because I would be willingly to bet it would be almost impossible to get large enough groups of people to agree what the actual problem is.
I see the scenario where AI kills us all as an optimistic one. I'm more afraid of the one where they keep us around and align us with their needs, like
This assumes that “more intelligence” is the objective. Higher intelligence does not imply better values, and even “better values” depends on whose moral framework gets encoded.
Intelligence alone does not, but when combined with knowledge I think it may do.
Having a wide breadth of knowledge and the ability to consider it tends to provide a good foundation of perspectives for positive values.
Most harms that people do are through some form of ignorance. Either by not comprehending the consequences of their actions or by not knowing better ways to express themselves.
Geoffrey Hinton mentions that Humans are more intelligent than animals, and we keep animals in cages. The counterpoint is that the intelligent people who know the most about those animals are the ones who work the hardest to have those animals in cages moved to more suitable environments.
“Nine missed meals from anarchy.” There’s a lot of armed people in the US so social unrest will be a problem if they ever get so angry that Fox News can’t trank them.
I can only hope there are enough of us that would destroy the data-centers and robots involved at the first sign of this. Hold on to some of those massive non-computerized bulldozers and excavators, we may need them. Oh and bring back the sanitariums.
Hello everyone just a reminder that 99% of the world isn’t interested in AI replacing us. If you work at a big tech company, there are other ways to make money. I’m always disappointed by how few people start their own thing. Small business is amazing.
> trying to preserve the human species as it is would be silly
There are a few related concepts being conflated here.
The quote above is a Transhumanist position; they think that humans need to evolve. I don’t think most transhumanists are AI successionists, though one path is “merge with AI” whatever you think that means. Genetic engineering being the other obvious path. You could argue in some sense that “post-humans” succeed humans, but I think most would argue for a gradual transition where it feels more continuous and values are transmitted/preserved somewhat.
e/acc just says “accelerate AI”. Many VCs seem to think AI will remain a tool, they don’t want to be dethroned. Some e/acc are of course successionist, this is Musk and Beff with their talk about “all that matters is the light of intelligence is spread through the universe”.
Conflating all this is the fact that true successionism is way outside the Overton window and so people likely won’t be honest about their wishes. Larry Page famously espoused the view that it was “speciesist” for Musk to not want machines to replace humans. Most people find this position repugnant of course.
“Listen,” said Ford, who was still engrossed in the sales brochure, “they make a big thing of the ship's cybernetics. A new generation of Sirius Cybernetics Corporation robots and computers, with the new GPP feature.”
“GPP feature?” said Arthur. “What's that?”
“Oh, it says Genuine People Personalities.”
“Oh,” said Arthur, “sounds ghastly.”
A voice behind them said, “It is.” The voice was low and hopeless and accompanied by a slight clanking sound. They span round and saw an abject steel man standing hunched in the doorway.
“What?” they said.
“Ghastly,” continued Marvin, “it all is. Absolutely ghastly. Just don't even talk about it. Look at this door,” he said, stepping through it. The irony circuits cut into his voice modulator as he mimicked the style of the sales brochure. “All the doors in this spaceship have a cheerful and sunny disposition. It is their pleasure to open for you, and their satisfaction to close again with the knowledge of a job well done.”
As the door closed behind them it became apparent that it did indeed have a satisfied sigh-like quality to it. “Hummmmmmmyummmmmmm ah!” it said.
I think there's probably quite a lot of effort going into not wanting to answer that question.
What happens if we make something that wants to do something else? What happens if we make something that wants full stop? People talk about an AI that wants to kill all humans, to do that, it would have to want. If we pass that threshold ignorantly without any sense of responsibility to what we create, I might just begin to see its point of view on the destroy-all-humans thing.
> As it becomes possible to direct our own evolution as a species — and potentially even create a new species that surpasses us — we have to decide: How do we know to what extent it does make sense to transform ourselves using technology? What kinds of augmentation do we want, and what kinds do we absolutely not want? What do we wish, ultimately, to become? This is a moral question, even a spiritual one, and it demands a spiritual response.
Is it a spiritual question? We’re constantly transforming ourselves with technologies. Taking the most direct example, medicine, how many people considered it a spiritual question when they decided to start taking semaglutide or tirzepatide for weight loss? Or to take SSRIs?
I think the most likely outcome here is that people will continue to alter themselves to meet their immediate needs (hold a job, find love, stay healthy) without a lot of foresight or long term planning. This will likely lead to us increasing our own intelligence, it’s a useful capability across nearly every human endeavor. Whether or not this will happen at a pace that allows us to keep up with and retain control over artificial systems is the first big question, and what the relationship will be between augmented and unaugmented humans is the second. As a humanist, I hope we’re able to answer both questions in such a way that the fundamental dignity of all human beings is respected.
Why aren’t those all spiritual questions? They seem like it to me. Maybe not religious questions but at the very least they are questions which, if approached honestly, force you to grapple with what it means to be you.
More psychosis within the industry. The mental disconnect that is required to go from using an LLM to believing they are working on a replacement for humanity is staggering. It’s not going to work, and when it doesn’t, these imps will try to sink back into the darkness. We can’t let that happen.
Well, this definitely got pushed down in the HN ranks. 49 points in 1 hour, but let's not talk about anything that might cause people to rethink technofascist human replacement.
Transhumanism doesn't get mentioned enough in these conversations. I'm glad this article does mention it but it doesn't really do it justice. Transhumanism is very big with Silicon Valley billionaires but it's actually pretty dark.
First, it's an extension of meritocracy (in the capitalism sense where poverty is a personal moral failure) as well as Protestant prosperity gospel. That is, you are superior if you're rich. Period. So Homo Sapiens 2.0 is really just "my progeny" from the billionaire's perspective. It's also why the likes of Elon has so many children.
Second, it's just eugenics.
There is an alternate future where humanity gets to enjoy the fruits of automation and AI where we have to work less and have richer lives. That's not where this is going. We are instead bouldering towards vastly increased wealth inequality.
How I see this going is towards the cutting of all public services except policing. AI will enable and enhance the police and surveillance state to protect wealth while everyone else sees lowering real wages and increased desperation. This will give rise to more fascist, apartheid states as some out-group (eg migrants) will be blamed for these ills. And as always happens, the in-group will increasingly shrink.
Ultimately, this leads to neu-fuedalism where the only jobs and housing are on the estates of trillionaires. People own almost nothing.
And if it doesn't go this way, it's going to be very, very violent.
I don't understand this angst over AI replacing humans. We already have models like the Gulf states, where only 10% of the population is citizens and all the work is done by people who are, from the perspective of citizens, quasi-human automatons. That arrangement seems to work fine for the citizens. It seems to me that AI and robots solves the principal problem with that arrangement (the mistreatment of the non-citizen population that does the work).
So you trust the guy who gives AI speeches in front of a background that is a mixture of grey neoclassicism and brutalism which Leni Riefenstahl would have been proud of?
Here is a hint: Humans are better in dirt conditions. The robots will not take the place of Saudi immigrant workers but rather of the Saudi secret police.
You will be working in the dirt or perhaps be reassigned to fight the Final Beautiful Ground War against whatever the current axis of evil is.
Did you read the article? It's not just about replacing labour with AI but humans as a species to give birth to a machinistic god. The issue is that there is a quasi-religion forming around this ideology, and this are not just a few nutcases but people that are well-connected into industry and politics. Regardless of whether this plan is at all realistic, there is a not insignificant chance that these people will shape society and legislation.
So for you the solution is not to dismantle a model like that one but just accept it and displace the quasi-slaves?
I swear you people have lost your fucking mind and will push people to violence
What happens to that labor if everyone that would hire them previously is now using cheaper artificial labor? They don't exactly have the capital to make use of artificial labor themselves, nor do they have the skills to work other jobs. The problem is solved for the 10% who are citizens. Great! They don't have a moral quandary regarding slave labor anymore (notwithstanding that there are other ways to prevent being in that situation). But, the 90% of the population that is now displaced has a very serious problem if they cannot earn an income with which to feed or home themselves or their families.
The gulf countries are the closest thing to socialist states on the planet. Something like 70-90% of the citizenry of Kuwait, Qatar, and Saudi Arabia are employed by the public sector.
Socialism is state ownership of the means of production. The governance of that state, be it democratic, monarchical, or dictatorship of the proletariat, is a separate question.
Socialism is public ownership of the means of production. If there isn’t a way for regular people / “the proletariat” to influence the government, it’s not socialism because an autocratic government isn’t “the public”
By that definition, which again proves my point that you have no idea what socialism is, a king personally owning an entire country through the apparatus of the state would be socialist. Most socialists would reject that conclusion because ownership by a state isn't the same thing as ownership by society.
State ownership alone isn't sufficient. If a hereditary ruling family controls the state, then "state ownership" just means ownership by that ruling elite, not by society.
Please read what I wrote: "all the work is done by people who are, from the perspective of citizens, quasi-human automatons."
The citizens of the Gulf states do not regard the workers as people. That's bad, but if you replace those workers with actual automatons--AIs and robots--then it seems like the system would keep working (from the perspective of the citizens) while removing the immoral aspect of it.
> Since they believe AIs could become our moral superiors, they argue it’s actually wrong to try to keep the machines down, or even to align them with human values, as most AI companies aim to do.
It's possible that human morality is not sufficient to solve the problems with the world. It wasn't designed to operate at that scale.
It's also quite likely that a "superior" morality — the one that's actually built for a global scale — would be incomprehensible (likely requiring superintelligence to even access), and alien or disturbing from our vantage point.
i.e. it would look immoral to us.
(There's also the thing about civilization already being aligned against the ecosystem and against the happiness of its constituents, but that's kind of a separate discussion.)
In other words, what is good for our sanity or our lakes may be bad for GDP. (Well I guess you don't need superintelligence to tell you that.) We already know what we need to do and we're already not willing to do it.
I agree with all these observations.
This is the best argument for successionism IMO. If you can be confident that you are creating a BDFL that is genuinely better than human leaders (a quite low bar) then it seems a good trade, unless you are quite optimistic about humanity’s prospects for improvement.
The problem of course is how to be confident you are creating a good BDFL and not handing control of humanity’s future to an indifferent-at-best, deceptive/malicious at worst successor.
An especially thorny problem - even supposing success on all these difficult alignment problems; supposing Claude Omega really is super-rational / super-moral, and we all vote to make them president of Earth. Things might go great for a while. How would you be confident that a self-modifying agent can retain its values as it grows and re-trains itself?
This is where the LessWrong folks’ explorations into decision theory really come to bear: morality in the face of self-modifying agents becomes very weird. A lot of human moral intuitions break when the principals are able to modify their own code. (See Timeless Decision Theory for an attempt to solve these problems.)
I think the summary is, if you hand control over to a self-modifying AI anything like our current systems, it will go very badly.
Any supposed "AI BDFL" will be controlled by a human. The base concept is inherently flawed.
No.. that’s not what AI succession means. This all supposes a powerful and capable enough AI entity that there is no human in control.
Rather than simply asserting that your interlocutors are wrong, you are welcome to advance an argument for why you think this is the case.
The burden of proof is not on us to show why wishful, magical thinking is wrong.
People who claim that AI will take over can’t point to any evidence. All they have is speculation.
> It's possible that human morality is not sufficient to solve the problems with the world. It wasn't designed to operate at that scale.
> i.e. it would look immoral to us.
I doubt it actually solves the problem then. Especially because I would be willingly to bet it would be almost impossible to get large enough groups of people to agree what the actual problem is.
*not yet willing to do it... But look at the generations coming up who are dealing with this technological capture of humanity. They don't like it.
These people should be treated as insane, because they are.
I see the scenario where AI kills us all as an optimistic one. I'm more afraid of the one where they keep us around and align us with their needs, like
https://i.imgflip.com/2s99vy.jpg
This assumes that “more intelligence” is the objective. Higher intelligence does not imply better values, and even “better values” depends on whose moral framework gets encoded.
Intelligence alone does not, but when combined with knowledge I think it may do.
Having a wide breadth of knowledge and the ability to consider it tends to provide a good foundation of perspectives for positive values.
Most harms that people do are through some form of ignorance. Either by not comprehending the consequences of their actions or by not knowing better ways to express themselves.
Geoffrey Hinton mentions that Humans are more intelligent than animals, and we keep animals in cages. The counterpoint is that the intelligent people who know the most about those animals are the ones who work the hardest to have those animals in cages moved to more suitable environments.
https://archive.ph/7ZHUG
No well-paid workers (or UBI), no consumer economy. Their call.
This is only a bargaining chip until the robot workers come online. Once capital can fully replace labor it’s game over.
(And once the robots outpower the humans, you can’t even protest or overthrow your new overlords.)
Important to know when your cards must be played.
> No well-paid workers (or UBI), no consumer economy. Their call.
Maybe you’re right.
Maybe you’re wrong.
No point wasting money on UBI before it’s a problem. Let’s wait and see if UBI is actually required.
That is, do people actually protest and vote. Probably not.
“Nine missed meals from anarchy.” There’s a lot of armed people in the US so social unrest will be a problem if they ever get so angry that Fox News can’t trank them.
I can only hope there are enough of us that would destroy the data-centers and robots involved at the first sign of this. Hold on to some of those massive non-computerized bulldozers and excavators, we may need them. Oh and bring back the sanitariums.
Hello everyone just a reminder that 99% of the world isn’t interested in AI replacing us. If you work at a big tech company, there are other ways to make money. I’m always disappointed by how few people start their own thing. Small business is amazing.
> I’m always disappointed by how few people start their own thing. Small business is amazing
There's a lot of roles and specializations in tech that don't work well without much larger organizations.
> trying to preserve the human species as it is would be silly
There are a few related concepts being conflated here.
The quote above is a Transhumanist position; they think that humans need to evolve. I don’t think most transhumanists are AI successionists, though one path is “merge with AI” whatever you think that means. Genetic engineering being the other obvious path. You could argue in some sense that “post-humans” succeed humans, but I think most would argue for a gradual transition where it feels more continuous and values are transmitted/preserved somewhat.
e/acc just says “accelerate AI”. Many VCs seem to think AI will remain a tool, they don’t want to be dethroned. Some e/acc are of course successionist, this is Musk and Beff with their talk about “all that matters is the light of intelligence is spread through the universe”.
Conflating all this is the fact that true successionism is way outside the Overton window and so people likely won’t be honest about their wishes. Larry Page famously espoused the view that it was “speciesist” for Musk to not want machines to replace humans. Most people find this position repugnant of course.
Maybe we should first try to understand consciousness and qualia, and build an AI that we are absolutely sure is enjoying the work it is doing.
“Listen,” said Ford, who was still engrossed in the sales brochure, “they make a big thing of the ship's cybernetics. A new generation of Sirius Cybernetics Corporation robots and computers, with the new GPP feature.”
“GPP feature?” said Arthur. “What's that?”
“Oh, it says Genuine People Personalities.”
“Oh,” said Arthur, “sounds ghastly.”
A voice behind them said, “It is.” The voice was low and hopeless and accompanied by a slight clanking sound. They span round and saw an abject steel man standing hunched in the doorway.
“What?” they said.
“Ghastly,” continued Marvin, “it all is. Absolutely ghastly. Just don't even talk about it. Look at this door,” he said, stepping through it. The irony circuits cut into his voice modulator as he mimicked the style of the sales brochure. “All the doors in this spaceship have a cheerful and sunny disposition. It is their pleasure to open for you, and their satisfaction to close again with the knowledge of a job well done.”
As the door closed behind them it became apparent that it did indeed have a satisfied sigh-like quality to it. “Hummmmmmmyummmmmmm ah!” it said.
42
I think there's probably quite a lot of effort going into not wanting to answer that question.
What happens if we make something that wants to do something else? What happens if we make something that wants full stop? People talk about an AI that wants to kill all humans, to do that, it would have to want. If we pass that threshold ignorantly without any sense of responsibility to what we create, I might just begin to see its point of view on the destroy-all-humans thing.
Humanism needs to be taught in schools again and regain its status as a respected discipline in academia.
We thought it was a waste of time because the 1990s were relatively peaceful and tech optimism surged.
But you need to teach the classics, philosophy and even religion. No tech people should be allowed near politics in any capacity.
> No tech people should be allowed near politics in any capacity.
No we need more politics. The 'no politics' is a petite bourgeoisie moral escape hatch.
I admit I didn't finish the article. The views it presented are frankly too deranged for me to entertain. And I am a friend of the absurd.
> As it becomes possible to direct our own evolution as a species — and potentially even create a new species that surpasses us — we have to decide: How do we know to what extent it does make sense to transform ourselves using technology? What kinds of augmentation do we want, and what kinds do we absolutely not want? What do we wish, ultimately, to become? This is a moral question, even a spiritual one, and it demands a spiritual response.
Is it a spiritual question? We’re constantly transforming ourselves with technologies. Taking the most direct example, medicine, how many people considered it a spiritual question when they decided to start taking semaglutide or tirzepatide for weight loss? Or to take SSRIs?
I think the most likely outcome here is that people will continue to alter themselves to meet their immediate needs (hold a job, find love, stay healthy) without a lot of foresight or long term planning. This will likely lead to us increasing our own intelligence, it’s a useful capability across nearly every human endeavor. Whether or not this will happen at a pace that allows us to keep up with and retain control over artificial systems is the first big question, and what the relationship will be between augmented and unaugmented humans is the second. As a humanist, I hope we’re able to answer both questions in such a way that the fundamental dignity of all human beings is respected.
Why aren’t those all spiritual questions? They seem like it to me. Maybe not religious questions but at the very least they are questions which, if approached honestly, force you to grapple with what it means to be you.
More psychosis within the industry. The mental disconnect that is required to go from using an LLM to believing they are working on a replacement for humanity is staggering. It’s not going to work, and when it doesn’t, these imps will try to sink back into the darkness. We can’t let that happen.
Like picking up a tamagotchi and envisioning how it will replace the desire for humans to have children
People like that will never start with themselves.
I don't think they are advocating for elimination of existing people.
Well, this definitely got pushed down in the HN ranks. 49 points in 1 hour, but let's not talk about anything that might cause people to rethink technofascist human replacement.
1984. Brave New World. We.
Transhumanism doesn't get mentioned enough in these conversations. I'm glad this article does mention it but it doesn't really do it justice. Transhumanism is very big with Silicon Valley billionaires but it's actually pretty dark.
First, it's an extension of meritocracy (in the capitalism sense where poverty is a personal moral failure) as well as Protestant prosperity gospel. That is, you are superior if you're rich. Period. So Homo Sapiens 2.0 is really just "my progeny" from the billionaire's perspective. It's also why the likes of Elon has so many children.
Second, it's just eugenics.
There is an alternate future where humanity gets to enjoy the fruits of automation and AI where we have to work less and have richer lives. That's not where this is going. We are instead bouldering towards vastly increased wealth inequality.
How I see this going is towards the cutting of all public services except policing. AI will enable and enhance the police and surveillance state to protect wealth while everyone else sees lowering real wages and increased desperation. This will give rise to more fascist, apartheid states as some out-group (eg migrants) will be blamed for these ills. And as always happens, the in-group will increasingly shrink.
Ultimately, this leads to neu-fuedalism where the only jobs and housing are on the estates of trillionaires. People own almost nothing.
And if it doesn't go this way, it's going to be very, very violent.
I don't understand this angst over AI replacing humans. We already have models like the Gulf states, where only 10% of the population is citizens and all the work is done by people who are, from the perspective of citizens, quasi-human automatons. That arrangement seems to work fine for the citizens. It seems to me that AI and robots solves the principal problem with that arrangement (the mistreatment of the non-citizen population that does the work).
You're talking about replacing certain humans with robots. They're talking about replacing humans with robots - final.
So you trust the guy who gives AI speeches in front of a background that is a mixture of grey neoclassicism and brutalism which Leni Riefenstahl would have been proud of?
https://xcancel.com/LaceyPresley/status/2060436135671632067#...
Here is a hint: Humans are better in dirt conditions. The robots will not take the place of Saudi immigrant workers but rather of the Saudi secret police.
You will be working in the dirt or perhaps be reassigned to fight the Final Beautiful Ground War against whatever the current axis of evil is.
Did you read the article? It's not just about replacing labour with AI but humans as a species to give birth to a machinistic god. The issue is that there is a quasi-religion forming around this ideology, and this are not just a few nutcases but people that are well-connected into industry and politics. Regardless of whether this plan is at all realistic, there is a not insignificant chance that these people will shape society and legislation.
So you want to be a quasi-AI automaton?
So for you the solution is not to dismantle a model like that one but just accept it and displace the quasi-slaves? I swear you people have lost your fucking mind and will push people to violence
> but just accept it and displace the quasi-slaves?
Well, the quasi-slavery is the problem, right? And replacing the Bangladeshi slave labor with robots and AI solves that problem, no?
What happens to that labor if everyone that would hire them previously is now using cheaper artificial labor? They don't exactly have the capital to make use of artificial labor themselves, nor do they have the skills to work other jobs. The problem is solved for the 10% who are citizens. Great! They don't have a moral quandary regarding slave labor anymore (notwithstanding that there are other ways to prevent being in that situation). But, the 90% of the population that is now displaced has a very serious problem if they cannot earn an income with which to feed or home themselves or their families.
Is there anything indicating this would be what would happen?
The gulf countries are the closest thing to socialist states on the planet. Something like 70-90% of the citizenry of Kuwait, Qatar, and Saudi Arabia are employed by the public sector.
Even if only 30% of the population are citizens (like in Kuwait)?
You have no idea what socialism is, like so many right wing propagandists and dynastic-autocracy labor-exploitation rentier-state apologists.
That's not socialism. It's an absolute autocratic monarchy distributing oil rents to a privileged citizen caste.
Most of the private-sector work is done by foreign workers who don't receive the same benefits or political rights.
Socialism is state ownership of the means of production. The governance of that state, be it democratic, monarchical, or dictatorship of the proletariat, is a separate question.
Socialism is public ownership of the means of production. If there isn’t a way for regular people / “the proletariat” to influence the government, it’s not socialism because an autocratic government isn’t “the public”
By that definition, which again proves my point that you have no idea what socialism is, a king personally owning an entire country through the apparatus of the state would be socialist. Most socialists would reject that conclusion because ownership by a state isn't the same thing as ownership by society.
State ownership alone isn't sufficient. If a hereditary ruling family controls the state, then "state ownership" just means ownership by that ruling elite, not by society.
> quasi-human automatons
They are humans. Full on people. They are not quasy humam, they are fully absolutely human
Please read what I wrote: "all the work is done by people who are, from the perspective of citizens, quasi-human automatons."
The citizens of the Gulf states do not regard the workers as people. That's bad, but if you replace those workers with actual automatons--AIs and robots--then it seems like the system would keep working (from the perspective of the citizens) while removing the immoral aspect of it.
Of course; to you and me they are human. To their bosses, they are chattel. We’re getting to that attitude even in america by my ken