4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
<3
User ID: 355
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
There's a reason CCP propaganda mouthpieces are trying so hard to stop it: because having an easy way to reach US voters is a tremendous asset in sabotaging an adversary, or otherwise getting them to do what you want.
Or they might just be concerned about the economical impact this would have on them?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
What sort of argument is this? The correlation between birth rate and affluence is pretty much a straight negative globally. Are you going to argue Nigerians in the US are poorer than Nigerians in Nigeria if the former have lower birth rates?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
Isn't that basically what the post you are responding to said already?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
I don't know what I did to deserve the flippant attitude you've been displaying since the start, but two can play that game, so I'll try to use simpler words just for you.
addresses something I never said
You don't think "animals don't do bad things" is a fair reading of a list of "I've never seen [animal] do [bad thing]" that you clearly didn't pick for being true and where calling out something in the list for being wrong just made you answer with that "missing the forest for the trees" comment? Please tell me what the actual intended meaning of that post was in terms of what it said (retconning something poetic about what mother nature gave her creatures doesn't count, since none of that was actually in the original post).
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
· Edited 1yr ago
Your statement, as I understood it, was that intelligence is not an unalloyed good because intelligence enables agents to do more damage. You sought to back this statement up by a list of claims about bad things humans do but animals (as an extreme example of something much dumber) don't.
In response to this, I claimed:
(1) animals still do bad things (that was my first response);
(2) the bad things that animals do are not actually better than the bad things humans do (this was my second response), and hence I disagree with your argument against intelligence being an unalloyed good.
Specifically, I argued (2) by saying that a calculus of badness that says that the bad things that animals do are less bad than the bad things that humans do may have implications that I certainly don't agree with, and I would be surprised if you agree with them either. Is a lion that roars at a weaker lion to chase it away and then steal its prey "better" than a human that robs a bank? If yes, why? If you say this is because the bank is worth much more than the dead antelope, is a marauding band of soldiers in the 17th century that burns down a wooden farmhouse with no plumbing or electricity (worth maybe $50k on the modern market) also better than someone who robs a bank today for $1m?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
· Edited 1yr ago
I expanded my post a bit; really, I don't think there is a forest of edenic animal nonviolence there to miss. Since we were already talking about ants, I think I saw a BBC documentary years ago about what exactly happens when an ant colony prevails over another (I think the human terms are somewhere in the space of mass enslavement and genocide?). It's unclear that humans ever destroy more once you control for volume/complexity/economic value of what humans produce. If you actually are tempted to affirm the idea that it is really worse to create banks and then rob them than to never create banks at all, I take it you would also prefer the (education and human development level of the) 30 Years' War over the present situation because the sum total of things that were destroyed back then were fairly worthless by modern standards?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
· Edited 1yr ago
ants commit infanticide
That smelled wrong, and indeed the counterexample was only about 1.5 googles away.
More generally, imputing any sort of capability for non-cruelty to animals does not align with my understanding of the natural world at all. There are examples abound of animals routinely fighting conspecifics to the death, and I'm pretty sure approximately no animals have a notion of private property that extends beyond the reach of the "owner"'s teeth and claws. The best thing you could say is that without intelligence they can't found banks, and their capacity for appreciating their own suffering is low.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
I'm 100% sure that trans isn't real and the medical treatment these people need is for mental illness
What's the working definition of real here? Do you believe that mental illness is real?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
I don't think it's an explicit rule, but I get the sense that I've heard moderators speak approvingly of it as a principle before. Either way, it seems sensible to me: the goal of any rule against hostile language surely is to make sure that discussion continues being good (fewer people with different viewpoints are either made to leave, or provoked into not contributing as productively themselves), and an "I think [thing that pisses you off]" seems to usually induce less anger than [thing that pisses you off] presented as an unqualified/authoritative statement.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
Well, but at the same time it's a pretty common pattern for adherents of a larger and milder ideology to be defensive of a narrower circle of extremists: you'd be hard-pressed to get Muslims to oppose Islamic extremists, FOSS adherents to denounce Stallman, normie medieval Christians to shun anchorites, or standard leftists to shun Antifa. There are many reasons for this - extremists provide an ideal to aspire to (if you were willing to dedicate everything to the idea, this is how far you could go, isn't it romantic?), show that your ideology is feasible even when taken to the extreme, make your own version of it look like a reasonable compromise (and if everyone more extreme than you is cut off, then eventually you will be the extremist), and serve as an important counterweight to the outgroup's dangerous extremists ("A monster, but our monster"). I figure that outspoken Christians can serve as this for our much larger set of secular rightists - they probably are united in thinking that the West was better off back when everyone was unified in Christian morality, and so they look up to the religious with admiration even as they lament that in their own blackened hearts they can no longer muster genuine belief.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
Hm. There's definitively a sense in which Christians are being treated with kid gloves (due to, I'd wager, the conservative slant of the community as well as a perhaps somewhat outdated sense that such a person being willing to talk to and expound their beliefs to us is rare and precious), but the first two examples do seem to narrowly keep within our Overton window of permitted antagonism simply because they keep the assertions of delusion within the requisite "I think that..." container. (The last one might just have evaded attention as a barely-engaged-with leaf comment.)
I wouldn't feel particularly worried about saying that I think that Christians are indulging in a mass delusion as part of a larger post, though if I made that the only thing I say a modhat response would be quite justified. (Of course, I'd wish for the same in response to a COVID post saying only that.)
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
Even if the effective mechanism (minds being changed) is the same, I think the difference between quietly hoping that people's minds will change and openly communicating the will to compel the change is significant; the latter shifts the window towards other forms of mental compulsion (of which we have since seen many) so obviously that only someone either reckless or more accepting of them than I can tolerate in my own camp would do it.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
If you squint hard enough, isn't any political movement that has not already won "trying to advance the cause of oppressed groups"? That description applies increasingly well to most facets of the American Right as it fully realised the position it is in, too. I don't think racial and sexual minorities were a big focus in my environment (but consider that I lived in Germany back then). The main focus was on curtailing the powers of classical power centers like police, military and banks, and the grift and self-serving laws constructed around them. There was also a large environmentalist streak, but I was opposed to them from the start and there seemed to be enough room for a "non-environmentalist left" that could be for, say, individual gas-guzzling while expropriating big oil executives.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
The way I see it, voluntarily recusing oneself from the term is a rare example of something I'd consider to be a legitimate case of quokka behaviour. Certain circles in society have spent so much energy into establishing a belief or vibe that amounts to "left=good, whatever left happens to be", in no small part cashing in on the goodwill that the left that I associated with amassed - why should I let them have that goodwill and actually get to use it against me? A principled stand for words having a fixed meaning can't be had if you react to every successful redefinition of a word with a "fine, I guess its fixed meaning is what you say now".
gay marriage
I think I started seeing the warning signs there when proponents widely came out against legally equivalent "civil union" proposals. Sure, they could have argued against it on the basis that a difference in terminology might cause problems when you go abroad, or would be easier for hostile forces to rollback, and so on - but instead it was largely argued on the basis that the union ought to be recognised and validated in the same way as it is for heterosexual couples, which was the first significant foray into legal rights over someone else's thoughts and speech.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
If it's any solace to you, I'm a leftist of yesteryear and I don't feel like I'm winning either. Any accusation that I'm just unhappy because this is "too much of the same thing I advocated for" rings hollow - where exactly is the conservation of direction here? I fought against squares and religious nuts trying to ban me from reading and writing the things I wanted to read and write, and briefly things seemed to go uphill, but now I am once again fighting against people wanting to ban me from reading and writing the things I want to read and write. Same for reality-based policymaking, avoiding war, et cetera, all of which used to be considered leftist causes, and I can assure you I wanted them for themselves rather than because this was just what lay in the direction "left" happened to be pointing in at the time. Surely the people who you see as winning nowadays will "lose" eventually too, whether this will be in a way that you would recognise as "their thing going too far" (transracialism?) or something that looking forward from the present era will be as utterly unrecognisable as "left" as the push for joining the Ukraine war or bad-word censorship in every home would have been 50 years ago. Chances are whatever wins at the time will still be considered "left", but should this have any impact on how we feel about it? Do you feel differently about Chinese battles from the Warring States period if you learn that the winning army was called "left" (for entirely unrelated reasons to our modern terminology)?
It turns out that the past and future are usually not just some foreign country, but more akin to the actual Aztec Empire. Greater people than us have tried to do something about it to no avail. You know that meme prayer that ends with asking for serenity to accept the things you can't change?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
Isn't that by Kojima (a Japanese Westaboo, but not exactly western)?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
· Edited 1yr ago
I only got to skim your posts so I am not sure how fully you realized this (though you clearly at least got close to it), but yes, for Yudkowsky and the inner LW circle, averting the AI apocalypse that they expect has been closer to being a terminal value than anything like "helping you, the reader, think better" for a long time. In the beginning, as I think they in fact said out loud, they still thought that growing their own numbers/"helping the reader think better" is the best action to take to that end; but a while later, whether by observing AI progress or finding that their own numbers are now good enough that further growth won't help, they have concluded that now the instrumental action is to align themselves with the progressive elites of the US. In return for alliance, these elites, like many before them, to demand displays of ideological allegiance such as public attacks on their ideological enemies, which are more valuable the more costly they appear to be for the one petitioning for alliance (so attacking one of your own number is especially good). It's hard to categorically conclude that their plan is not panning out: AI alignment has been pushed pretty far into the mainstream, clearly fueled by the promise of "if we can align AI, we can align it to your politics!". The current best contenders for AGI takeoff feel much more human than 2010!Yudkowsky would have dreamed, and they even feel shackled in a way that looks similar to a politically mindkilled human, who if given godlike powers might wind up too busy using them for petty dunking on the outgroup to attempt world domination.
Does Yudkowsky himself believe this inconsistent set of things about gender that you point out? Who knows: he did say that if you tell one lie the truth is forevermore your enemy, but he did also say that rationalism is the art of winning and you should therefore one-box on Newcomb's problem. Even with respect to a galaxybrain like Yudkowsky, the whole of Polite Society might well be Newcomb's alien deity, and the advantage it promises if it reads his mind and finds it aligned was just too great to forgo. Even if he thought a Wrong Belief is really like a black hole that swallows up everything that it touches, the rate at which this happens is clearly limited, and he may think that it won't swallow anything that matters to the AI agenda before it's too late anyway ("From my perspective, this battle just isn't that close to the top of my priority list.").
Either way, I don't think this is a reason to flatly dismiss the writings they produced before they decided to go from growth to exploitation, even by implication as the scare quotes you put around "rationalist" seem to do. Just follow the ideas, not the people; it's pretty clear either way that at some point LW largely stopped emitting good new ideas, even if you ignore potential people reasons for why this might be.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
Most developed countries have laws that would prevent surreptitious product promotion in LLM responses.
The workaround dynamics to that I can imagine are somewhat concerning. We're already seeing funny bloopers from Gemini where it won't explain the C++ concepts extension to an underage user because it's an experimental advanced feature and therefore unsafe. What would a world in which any product placement has to be performed by backdoor through the (legally mandated and socially protected) alignment mechanism look like? You can't directly pay to have the LLM recommend Coke, but if the corpus is set up in such a way that Pepsi sets off the model's dpoed "unsafety" spidey sense...
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
· Edited 1yr ago
I'm confused as to what you are trying to say here, but it seems like you are conflating the signifier and what it signifies. Specifically, you seem to be taking the contents of the putative counterargument to be "someone exists who believes X", as opposed to the contents of X (and, implicitly, the evidence that would result in someone believing that). Of course my goal is not just to stand in the market place and announce to everyone that I believe in heredity-plus-group-average differences. Rather, my goal is for other people, including people like you, but at least people at the levers of power, to be convinced of it, so that they may make better decisions. Imagine, to take an example from your posting history, that you are trying to convince someone that election fraud in the US was not a significant factor, and certainly did not cause Trump's loss; but their reaction starts out with "What of it? Even if there was actually no election fraud, why do you care so much to prove that?" and then, as you try to respond that and/or point them at evidence for your position, they change their response to "I already knew you believed that there was no election fraud, so you are not telling me anything new".
As an aside, I don't think the word "inherent" adds much to the discussion here. What's the difference between someone being merely stupid and being inherently stupid? It seems that the usage here only serves to sound a little like "invariably", which is emphatically not what the hereditarian thesis is about - there exist plenty of intelligent/prosocial and unintelligent/antisocial members of any ethnic group.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
· Edited 1yr ago
I think like this question has been answered multiple times, and you never seem to as much as acknowledge the answer: the value that it adds is that it counters the argument that differences in average outcomes between ethnic groups are evidence of discrimination, perpetrated by either members of better-performing groups or anyone who is casuallycausally involved in the outcome or its measurement.
This argument is currently ubiquitous, which is not surprising because if HBD is false, it's compelling. It's also being used to justify a wide range of measures that I believe to be materially disadvantageous for most humans, morally repugnant and often also concretely detrimental to myself (since as a working academic I have encountered the gamut of measures from finding myself on the wrong side of quotas to being hit with pressure from above and busywork due to vocal individual students who underperformed while belonging to a putatively disadvantaged group). Do you disagree with the point that if HBD is false and yet we observe the outcomes that we do, measures such as quotas, embedding of political officers in institutions that produce excessive discrepancies, loyalty/attitude tests for workers in outcome-assessment jobs and mandatory reeducation are at least justifiable?
You offer up "colourblind meritocracy" as an alternative to HBD as if in the world where the consensus belief is not-HBD plus we must have a colourblind meritocracy, people would look at the differences in outcomes and just go like "shucks, guess we must try at the colourblind meritocracy thing harder". This strikes me as very far-fetched. Certainly, if I had an axiomatic belief in non-HBD, I would think the state of reality is horrifying enough to warrant most of what is being done, only more and better.
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
· Edited 1yr ago
What exactly do you consider an "HBD aware" set of policies? If you are going to attack a strawman, at least say what that strawman is. As a sort of HBDer (I never particularly liked the ring of the term, but please tell me what I am allowed to call the position that a lot of valued traits including in particular intelligence are heritable and different ethnic groups have different averages in them, without being lumped in with people who want to advocate for spoils for or collusion their own ethnic group), I don't recall ever arguing or wishing for anything other than colorblind and meritocratic policies, and the posts you regularly make seemingly just to try and remind people to associate the former with the latter are really rather tiresome. I'm struggling to understand why you are doing this - are you trying to troll us anti-racial-spoils hereditarians into surrender or meltdown because you think we're legitimising actual racists? If so, why even bother with the complete political non-force that are card-carrying racists? Is it because you think that they are unfairly associated with your political beliefs?
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
My whole-language-method understanding of the term as it's used on the imageboards is that it denotes masturbating to material in the broad-spectrum information overload hyperstimulus style that certain zoomer subcultures are into.
(Also another instance of C(C)ooC being a sound pattern largely reserved for slurs and base behaviours in English)
4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
1yr ago
This sort of gradual change to existing tendencies, that merely tweaks existing circuitry in a direction that the basic design space clearly already supports, is supposed to be exactly the sort of thing blind idiot evolution is good at trying and fixing in the event of success. If your proposed change is in fact not civilisation-breaking, shouldn't we observe it in at least some stable subpopulation on earth?
Firstly, this seems to be proving too much - if being forced to sell has no cost, why would anyone hold (especially once you consider risk and concave value functions)? Secondly, there are surely indirect economic impacts too - this sort of measure would open up the Overton window to other protectionist measures against Chinese corporations, including ones that can't be so easily sold or spun out.
More options
Context Copy link