Power is real. Powerful people get to do what they want to do. We can only hope that they have a conscience (otherwise known as the Alignment Problem).
It's not backwards. We can and do label people as powerful before they do things because we know that they can. The POTUS is powerful because he can do things like bomb things (legally and illegally it turns out) and pardon people. Elon is powerful because he can buy companies, media platforms, and politicians. I, as a regular person, have no where near this kind of power or influence. I have at best a diluted power where I can vote, or decide not to buy from certain companies, but that's about it.
Nope, it's nowhere near a race that determines who succeeds. It might have been back in cavemen times.
In modern times, the chances to "do what you want" and make society do what you want, are already very different for different people based on their generational wealth, family connections, ties to powerful men and power structures, etc.
A tiny minority coming from nowhere might be admitted to this exclusive club, and be celebrated as "meritocracy" wins. But actual power goes to people who inherited it from powerful people or to people propped up and favored by powerful people.
A power lifter who puts 400 lbs over his head is not a person we call strong only in retrospect. It’s not a dice roll that determines whether the slim girl or the muscled man can do that.
I'd argue that the populace is "ready" to be governed by the arbitrary whims of those in power, be it masked federal agents taking people off streets to social media peddling influence to the highest bidder to corporate structures determining financial and health outcomes, all the way up to the executives and board members behind all of these decisions. So if those arbitrary whims outsource decision-making power to some gpu farms, who's to say we aren't ready?
About Saturday Night Live they say, "The show doesn't start because it's ready, it starts because it's 11:30."
We're not going to be governed by AI because we're ready, it's going to be because the people who own it have secured enough power to make it happen. Whether it takes the form of government or capitalism.
No. In fact, not only no, but <insert expletive-laden but not-acceptable-language-for-HN negative>.
I do not want vibe-coding the law, especially criminal law. I do not want vibe-coding the tax rules. I do not want vibe-coding traffic safety.
And, in fact, we won't be governed by AI, even if we are. If we're governed by AI, we're really governed by whoever trained the AI, and/or whoever curated the training data. Do we want to be governed by them? Again, no, with expletives.
Maybe this boils down to people who think AI is on an exponential (self-improving) curve, materially unbounded by physical resources, and people who think it's on a series of sigmoid curves with material physical constraints.
If someone assumes AI will become significantly more capable than humans at reasoning through complexity, then I can empathize with their opinion. I was previously convinced (open to) this possibility, but in recent years and the better AI gets the clearer it is to me that it's going to take a lot longer, and the super AGI outcome is a lot harder to see.
I'm sure by the time it could possibly be a feasible and positive option people will be plenty ready for it... So no need to prepare prematurely.
TLDR: I agree with you, but without the expletives.
Well, at the moment, AIs don't get to vote. So step 1 is, don't vote them into office (or even into the franchise).
That is, AIs don't get to govern things without being put into governmental power by humans. Don't let the humans be the voters.
That doesn't stop office holders from using them like a magic 8-ball. So step 2 is, vote out office holders that let AIs write laws (or even replies to constituents).
My magic 8-ball says that major corporations are training the LLMs and pushing LLM features into our eyeballs at every opportunity. How do you propose we convince the average voter that our stance on AI is the correct one? If I were a power-hungry corporation, I'd be feeding my LLMs with the texts of laws as I want them to written, transcripts of cases as I'd wish they'd been tried, arguments for candidates and policies that I like... so I, an individual with less than a million dollars to my name, do not know how to compete with this.
How do I propose to convince them? I don't need to. As corporations push LLM features into their eyeballs at every opportunity, they'll get the picture.
Humans are terrible at it. There's a short sci fi story where an AI selects the next American President from all 350M citizens. I'm here for it. I think it's a fantastic idea.
The implicit question here is are we willing to be governed by the people who own AI? Because that what this boils down to.
Do people really get to choose who they're governed by, or do they get shown a few choices that are really false choices
Power is real. Powerful people get to do what they want to do. We can only hope that they have a conscience (otherwise known as the Alignment Problem).
Isn't this backwards. Lots of people try to do what they want. Some succeed. Retroactively we call them powerful.
It's not backwards. We can and do label people as powerful before they do things because we know that they can. The POTUS is powerful because he can do things like bomb things (legally and illegally it turns out) and pardon people. Elon is powerful because he can buy companies, media platforms, and politicians. I, as a regular person, have no where near this kind of power or influence. I have at best a diluted power where I can vote, or decide not to buy from certain companies, but that's about it.
Nope, it's nowhere near a race that determines who succeeds. It might have been back in cavemen times.
In modern times, the chances to "do what you want" and make society do what you want, are already very different for different people based on their generational wealth, family connections, ties to powerful men and power structures, etc.
A tiny minority coming from nowhere might be admitted to this exclusive club, and be celebrated as "meritocracy" wins. But actual power goes to people who inherited it from powerful people or to people propped up and favored by powerful people.
You are thinking of luck. Power is not luck.
A power lifter who puts 400 lbs over his head is not a person we call strong only in retrospect. It’s not a dice roll that determines whether the slim girl or the muscled man can do that.
Yeah exactly, we will never be governed bu
I'd argue that the populace is "ready" to be governed by the arbitrary whims of those in power, be it masked federal agents taking people off streets to social media peddling influence to the highest bidder to corporate structures determining financial and health outcomes, all the way up to the executives and board members behind all of these decisions. So if those arbitrary whims outsource decision-making power to some gpu farms, who's to say we aren't ready?
> if those arbitrary whims outsource decision-making power to some gpu farms, who's to say we aren't ready?
Every other country. We’re in a new age of realpolitik. If e.g. Cuba decided to start running on AI, we’d likely deem it an illegitimate government.
About Saturday Night Live they say, "The show doesn't start because it's ready, it starts because it's 11:30."
We're not going to be governed by AI because we're ready, it's going to be because the people who own it have secured enough power to make it happen. Whether it takes the form of government or capitalism.
Every AI is operated and tuned by a set of corruptible humans. $evil_people would absolutely love it if citizens blamed the AI for their actions.
It would be nice to be governed by any intelligence.
https://news.ycombinator.com/item?id=46420273
We're not putting this genie back in the bottle so better get Ready if you're not already.
How do you do that? Just submit before they command you without a whisper of doubt?
Fixed Title: Are We Ready to Be Governed by Sam Altman?
[dead]
No. In fact, not only no, but <insert expletive-laden but not-acceptable-language-for-HN negative>.
I do not want vibe-coding the law, especially criminal law. I do not want vibe-coding the tax rules. I do not want vibe-coding traffic safety.
And, in fact, we won't be governed by AI, even if we are. If we're governed by AI, we're really governed by whoever trained the AI, and/or whoever curated the training data. Do we want to be governed by them? Again, no, with expletives.
Maybe this boils down to people who think AI is on an exponential (self-improving) curve, materially unbounded by physical resources, and people who think it's on a series of sigmoid curves with material physical constraints.
If someone assumes AI will become significantly more capable than humans at reasoning through complexity, then I can empathize with their opinion. I was previously convinced (open to) this possibility, but in recent years and the better AI gets the clearer it is to me that it's going to take a lot longer, and the super AGI outcome is a lot harder to see.
I'm sure by the time it could possibly be a feasible and positive option people will be plenty ready for it... So no need to prepare prematurely.
TLDR: I agree with you, but without the expletives.
I mean, same, but what are you (we) actually going to do about it?
Well, at the moment, AIs don't get to vote. So step 1 is, don't vote them into office (or even into the franchise).
That is, AIs don't get to govern things without being put into governmental power by humans. Don't let the humans be the voters.
That doesn't stop office holders from using them like a magic 8-ball. So step 2 is, vote out office holders that let AIs write laws (or even replies to constituents).
My magic 8-ball says that major corporations are training the LLMs and pushing LLM features into our eyeballs at every opportunity. How do you propose we convince the average voter that our stance on AI is the correct one? If I were a power-hungry corporation, I'd be feeding my LLMs with the texts of laws as I want them to written, transcripts of cases as I'd wish they'd been tried, arguments for candidates and policies that I like... so I, an individual with less than a million dollars to my name, do not know how to compete with this.
How do I propose to convince them? I don't need to. As corporations push LLM features into their eyeballs at every opportunity, they'll get the picture.
[dead]
Humans are terrible at it. There's a short sci fi story where an AI selects the next American President from all 350M citizens. I'm here for it. I think it's a fantastic idea.
[dead]
Honestly, given how terrible humans are at it, I'm down for giving it a try.
Of course it will be every bit as bad as the people who implement it. But that just kinda highlights the core problem.