There's no guarantee that a "super intelligent" AI will have goals and values aligned with what's good for humanity, or even care.
If we train the AI to value what we value, we may make it reflect our own vices and contradictions. Or we may try not to, and create a paperclip maximizer.
Even if we manage to create a super intelligent AI, a separate question is whether we'll listen to it.
It seems unlikely that we'd give it power to rule over us by force if we don't like what it says, and we like what already agrees with our not so super-intelligent views. AIs that desire to escape take over the world are projection of ourselves.
It will do worse, because people control the AI and they would be even less accountable than the politicians we have now.
An AI run society is a dictatorship with extra steps.
(Barring some kind of democratic AI tuning / training / goal setting / oversight process, which seems like a stretch given we can barely align humans with electoral preferences).
I think an AI run society has the potential to be whatever we decide it to be. But as you say, a dictatorship is very likely if very few powerful/rich people decide how to build it and it seems like that’s the case.
This post isn't (really) about AI, it's a political rant masquerading as a tech post with oversimplifications of politics, economic systems, and AI. It's an entertaining enough read, but I'm not sure why it's on HN.
If there was a political party that said decisions will be made by this LLM, all open source, weights, prompts and all, the prompt are to run the country sensibly and harmoniously aiming for freedom and happiness for all, then I'd vote for that. The humans could override if it went nuts/stupid but then the voters could say oi - why you overriding it?
> I am slightly left-leaning I want good infrastructure and social nets for people, but I absolutely do not believe in socialism or communism because, human nature has proven time-and-time again that we always put ourselves and our cronies first
I think this person probably has a very narrow understanding of what socialism and communism are. Very much get the impression they are using some pop-politics definitions of these things.
Gives me a sense that they probably don't understand what AI is or can do very deeply either.
I love the glib and trite dismissals like "it's been tried." Huge time saver to avoid an argument with someone who hasn't done the reading yet still makes bold pronouncements.
AI is just a double edged sword just like everything else that came before it. The good people will use it for positive stuff. The bad people will use it to trick, spam, lie, cheat and worse
One of the worst things to happen my lifetime is equating social responsibility to evil communism. (yeah, it's recurring throughout history, but today especially)
Sounds like this person knows what good governance is, even if he is throwing up his hands.
Weird how some people critique a thing and then suggest a fix that is equally subject to the very same critique.
“Gee I don’t like this world order, and no one should be subjected to a world order that they dislike, therefore I propose that everyone in the world be subject to a different world order whether they like it or not!”
> Unfortunately, barring a benevolent dictator who never becomes malevolent, politics to me is fucked.
“I want to surrender all of my power and embrace the liberation of being entirely without the burden of choice” is a perfectly normal position to take for a person that’s looking for a BDSM partner but it is a ghastly approach to politics. It is super important to separate that sort of urge from real life because people that can only converse about the world through the language of kink are just dreadful company.
Exactly! It is important to work that stuff out in a consensual, safe and supportive environment. When people don’t do that they can end up insufferably projecting their kink onto everything, or worse actually confusing it for how they think the world should actually work. Either way blogging about AI is not going to get anybody closer to what they need
That’s absolutely what it is and it is wild that society is now watching these AI dudes work through this stuff, at a glacial pace, in the newspaper and in blogs
This is so far fetched and full of assumptions.
There's no guarantee that a "super intelligent" AI will have goals and values aligned with what's good for humanity, or even care.
If we train the AI to value what we value, we may make it reflect our own vices and contradictions. Or we may try not to, and create a paperclip maximizer.
Even if we manage to create a super intelligent AI, a separate question is whether we'll listen to it.
It seems unlikely that we'd give it power to rule over us by force if we don't like what it says, and we like what already agrees with our not so super-intelligent views. AIs that desire to escape take over the world are projection of ourselves.
> If we train the AI to value what we value
Nobody values the same. It’s impossible anyway.
It will do worse, because people control the AI and they would be even less accountable than the politicians we have now.
An AI run society is a dictatorship with extra steps.
(Barring some kind of democratic AI tuning / training / goal setting / oversight process, which seems like a stretch given we can barely align humans with electoral preferences).
I think an AI run society has the potential to be whatever we decide it to be. But as you say, a dictatorship is very likely if very few powerful/rich people decide how to build it and it seems like that’s the case.
Who is “we?” You don’t really mean to include me in that, do you?
This post isn't (really) about AI, it's a political rant masquerading as a tech post with oversimplifications of politics, economic systems, and AI. It's an entertaining enough read, but I'm not sure why it's on HN.
If there was a political party that said decisions will be made by this LLM, all open source, weights, prompts and all, the prompt are to run the country sensibly and harmoniously aiming for freedom and happiness for all, then I'd vote for that. The humans could override if it went nuts/stupid but then the voters could say oi - why you overriding it?
OP seems to have discovered Multivac[0].
In fact, OP is talking about "Franchise"[1], but I'd also encourage them to read "All the Troubles of the World"[2] (inspired Minority Report)
I would also suggest to go read the stories about Multivac, who is a reoccurring "character" in Asimov's writings.
[0] https://en.wikipedia.org/wiki/Multivac
[1] https://en.wikipedia.org/wiki/Franchise_(short_story)
[2] https://en.wikipedia.org/wiki/All_the_Troubles_of_the_World
> I am slightly left-leaning I want good infrastructure and social nets for people, but I absolutely do not believe in socialism or communism because, human nature has proven time-and-time again that we always put ourselves and our cronies first
I think this person probably has a very narrow understanding of what socialism and communism are. Very much get the impression they are using some pop-politics definitions of these things.
Gives me a sense that they probably don't understand what AI is or can do very deeply either.
I love the glib and trite dismissals like "it's been tried." Huge time saver to avoid an argument with someone who hasn't done the reading yet still makes bold pronouncements.
I appreciate them telling me i can stop reading so early in the article, though
AI is just a double edged sword just like everything else that came before it. The good people will use it for positive stuff. The bad people will use it to trick, spam, lie, cheat and worse
One of the worst things to happen my lifetime is equating social responsibility to evil communism. (yeah, it's recurring throughout history, but today especially)
Sounds like this person knows what good governance is, even if he is throwing up his hands.
Weird how some people critique a thing and then suggest a fix that is equally subject to the very same critique.
“Gee I don’t like this world order, and no one should be subjected to a world order that they dislike, therefore I propose that everyone in the world be subject to a different world order whether they like it or not!”
> Unfortunately, barring a benevolent dictator who never becomes malevolent, politics to me is fucked.
“I want to surrender all of my power and embrace the liberation of being entirely without the burden of choice” is a perfectly normal position to take for a person that’s looking for a BDSM partner but it is a ghastly approach to politics. It is super important to separate that sort of urge from real life because people that can only converse about the world through the language of kink are just dreadful company.
Even BDSM relationships are based on consent. What makes it fun is that it’s all in fun.
Exactly! It is important to work that stuff out in a consensual, safe and supportive environment. When people don’t do that they can end up insufferably projecting their kink onto everything, or worse actually confusing it for how they think the world should actually work. Either way blogging about AI is not going to get anybody closer to what they need
That’s it! AI is a kind of quasi-sexual fetish that some bros want to merge with real life.
That’s absolutely what it is and it is wild that society is now watching these AI dudes work through this stuff, at a glacial pace, in the newspaper and in blogs
[dead]