OliveTapenade
No bio...
User ID: 1729
Man, the Chaser takes me back a bit. There are a lot of good teenage memories watching those guys.
Who are the serious right wing intellectuals at the moment, in your view? Hanania talks a lot about the right lacking human capital, and in this very post argues that it's less and less tenable to be an intelligent person on the right. Who would some counter-examples be?
It does not seem to me that positive change can be built on just flipping off all the bad people.
If nothing else, one way to offend bad people is to also be bad yourself. There has to be a positive vision somewhere. A politics of nothing but contempt and hatred is an inherently sterile politics.
I'm not sure that we shouldn't care about African conflicts, but that's beside the point. The point is that, for better or worse, Western media appears to disproportionately care about Gaza. The suffering in Gaza has received a great deal more media attention than comparable suffering in other places. Don't like Artsakh? Fine, then, take Rakhine state - over a third again as large as Gaza in terms of population, and subject to similar brutality. Don't like that one? South Sudan. And so on.
It is thus, I think, not true that there has some kind of conspiracy of silence around Gaza. On the contrary, if there's a conspiracy I feel like it was to bring more attention to Gaza, not less.
There are some understandable reasons for the greater focus on Gaza - it involves a close American ally, the history of decades of Palestine as a symbol for wider Arab nationalism, lots of Jews in Western countries who pay particularly close attention to Israel, and so on. I'm not saying it's a great mystery why Americans take much closer interest in Gaza than they do in the plight of the Rohingyas. I'm just saying that they do. Thus this:
The vast and dishonest Zionist campaign in media and astroturfed across the internet to pretend that nothing was happening in Gaza and if it was happening it was a good thing was the best thing that happened to Holocaust Denial since, well, the Holocaust.
This is just not true. There was a huge amount of media focus on and discussion of Gaza, much of it openly critical of Israel. The 'vast and dishonest campaign' you posit seems hallucinatory. As 2rafa and upsidedownmotter show, it's just, well, not a true description of Western reporting on Gaza.
Note that I am not claiming that Western reporting on Gaza is unbiased, or plainly representative of the facts, or anything like that. I am just claiming that firstly there's a lot of it and secondly much of it is critical of Israel, whether implicitly or explicitly, so much so that I don't think one can defensibly claim that events in Gaza have been either hushed up or whitewashed.
It seems to me that there are at least three separate things here, if we consider the human example.
-
The actual cause of a human's decision. This is often unconscious and not accurately known even by the person making the decision.
-
The reasons a person will tell you that they made a decision, whether before or after the decision itself. This is often an explanation or rationalisation for an action made after the decision was taken, for invisible type-1 reasons.
-
The action the person takes.
I would find it entirely unsurprising if you did a study with two groups, one of which you ask to make a decision, and the other of which you ask to explain the process by which they would make a decision and then subsequently make a decision, those two groups would show different decisions. Asking someone to reflect on a decision before they make it will influence their behaviour.
In the case of the LLMs with the thought boxes, my understanding was that we are interested in the LLM's 1, i.e. the actual reasons it takes particular actions, but that the box, at best, can only give you 2. (And just like a human's 2, the LLM's stated thought process is only unreliably connected, at best, to the actual decision-making process.)
I thought that what we were interested in was 1 - we want to know the real process so that we can shape or modify it to suit our needs. So I'm confused as to why, it seems to me, some commentators behave as if the thought box tells us anything relevant.
Oh, well, I don't disagree with that at all. The Nazis were horrific, cruel, and morally evil, but they are very far from unique. Even in their own day, in the 30s and 40s, there were not unrivalled by other governments and organisations, and in the history of the 20th century after them, there were plenty of governments that match the Nazis for horror.
I'd go so far as to say that the portrayal of the Nazis as unique and singular in their evil is actively bad for understanding them and for preventing similar evils occurring again in the future.
I'm actually quite skeptical that there is anything that can be meaningfully described as a thought process or reasoning going on when an LLM responds to a problem like this. It may well be that if an LLM produces a step-by-step summary of how to go about answering a question, it then produces a better answer to that question, but I don't understand how you can draw any conclusions about the LLM's 'reasoning', to the extent that such a thing even exists, from that summary.
Or, well, I presume that the point of the CoT summary is to give a indicative look at the process by which the LLM developed a different piece of content. Let's set aside words like 'thought' or 'reasoning' entirely and just talk about systems and processes. My confusion is that I don't see any mechanism by which the CoT summary would correspond to the subsequent process.
It seems to me that what the paper does is ask the LLM to produce a step-by-step set of instructions, and then ask the LLM to iterate on those instructions. LLMs can do that, and obviously if you change the set of instructions, the iteration on the instructions is different. That's perfectly intuitive. But how does any of that correspond to, well, the idea of thoughts in the LLM's mind? Or the process by which it produces text? How is that different to the rather banal observation that if you change the input, you change the output?
But if nobody focuses on or sacralises specific details about the camps, then what is the deception? What is it that people recognise they were deceived about and then respond by questioning the entire event? The broad-strokes, big-picture narrative of the Holocaust (i.e. the Nazis, during WWII, deliberately attempted to exterminate the entire Jewish population of Europe, and transported them to death camps where they massacred them) is undoubtedly true, and if we agree that nobody outside a few niche obsessives particularly cares about the details of those camps' operation, what is the deception?
I mean, I don't think I'd ever heard the word 'zyklon' in my life before I visited the Motte. I just wasn't told a specific story about how the Nazis murdered the Jews, beyond maybe a vague "they gassed them".
So I guess I'm completely unmoved by the idea that there might be a valid historical debate about the exact methods. Heck, the Nazis weren't the most scrupulously organised group in the world and the camps were mostly destroyed, so I would not be surprised if a range of techniques were used in different places. So I have no sense of there even being a tall tale.
What I remember from my childhood is that Hitler was bad, basically. I did units in school on the lead-up to WWI and then on the rise of fascism, but ironically I never actually did the wars. They weren't offered - we skipped from the assassination of Archduke Ferdinand to the Treaty of Versailles, and then spent a semester on Weimar Germany and Hitler's rise to power, and then we skipped the war itself and came back the next year with the WWII peace settlement and the beginning of the Cold War. Presumably it was felt that there was no particular need for kids to understand all the military maneuvers; and interested kids (like me) would just go to the library and read all the military history books and pore over the maps (which I did).
I don't remember ever doing anything about Stalin, since he leaves the picture fairly early in the Cold War. There was a very common unit in Australian high school history about the Russian Revolution, though, which I didn't do (I had a different course), but presumably would have covered that. So most of what I got was that Hitler is bad and Nazis are bad, and they came to power in such-and-such way, and these are the sorts of things we should look out for in case it happens again, but the blow-by-blow of the war itself was not covered. That also meant that the logistics of the Holocaust were skipped entirely.
It's not as though we did nothing about the Nazis and the Jews - I read The Diary of Anne Frank, and I remember watching and writing about the film Au revoir les enfants, though those were in English class and French class respectively, not history per se. But the emphasis in works like that is much more about the breakdown in social trust and solidarity, people informing on their neighbours, and so on. I don't mind that particularly, because I think that is in fact the most important lesson to take from the Holocaust.
Anyway, I suppose it's possible that if you were taught very specific details about the death camps and sacralised those details, then learning that they might be incorrect or debatable would be challenging? But that sounds like a very different form of education to the one I received, and to be honest one that seems to me to have quite strange priorities.
Nope. I remember going to a Holocaust museum at primary school and that never came up. This is the first I've ever heard of anything involving lampshades either. The focus was much more on the history of intolerance, hostility, and repression that led up to the Nazis deciding to just kill all the Jews. The gruesome details of how they were killed at the camps were not gone into, I would guess partly because it's not child-appropriate and partly because it's not actually a useful thing to know in terms of what they were trying to teach us.
I don't have much experience talking to Holocaust deniers outside of the Motte, since, well, this is the only place I've ever met any, but one of the things that surprises me about them is their obsession with quibbling what seem to me like trifling details. I don't particularly care about the exact methods by which the Nazis murdered millions of Jews. I care that they murdered millions of Jews. That's the morally significant part.
Insofar as I find other questions about the Holocaust historically or philosophically interesting, I think for me it's mostly causal stuff? The intentionalist/functionalist debates, for instance, strike me as both interesting and instructive, in terms of how atrocities come to occur. But details about ovens or showers or what have you strike me as being mostly of niche academic interest.
I believe most definitions of genocide include occupying a territory and forcing all of its inhabitants to leave.
My experience has been that normie conservatives basically all hold that Hitler and Stalin were equally or near-equally bad, and that fascism and communism are equally bad.
On the left people will absolutely argue that communism is good, and occasionally you get someone who will argue that Stalin, while very bad, wasn't as bad as Hitler, but that's as far as it seems to go to me. "Who was worse, Hitler or Stalin?" is a question random people debate in bars. The idea that they're both equally bad is so common that it's a Mitchell and Webb sketch.
Wait, where was there a vast media conspiracy to say that nothing bad was happening in Gaza?
That sounds like the opposite of how I remember the last year and a half. On the contrary, it seems to me that the mainstream media has been obsessed with Gaza in a manner totally disproportionate to its actual importance. There has been a constant feed of events from Gaza, especially those critical of Israel - I remember a few weeks when the media could not stop talking about one specific hospital building that the IDF attacked.
If you compare coverage of Gaza to, say, coverage of Artsakh, which happened around the same time as October 7 and was much more unquestionably a genocide, the difference is stark. I suspect Gaza is that way because firstly there's more direct American involvement with Israel, secondly there's a large constituency in Western countries that cares about it (i.e. Jews, who are both wealthier and more influential than the Armenian diaspora), thirdly pro-Palestine activism has been a cause of the left for decades so there's a pre-existing infrastructure, and fourthly Gaza is just ambiguous enough to be spicy. A more obvious or unambiguous genocide doesn't mobilise the existing political coalitions, one to defend and one to attack. It has to be in the just-right zone, bad enough to mobilise people against it, but not so bad that people won't defend it. Gaza is in the zone - just ambiguous enough to be one movie and two screens, just enough for "Israel is committing a genocide!" and "Israel is defending itself from murderous fanatics!" to be both more-or-less defensible claims.
I think when you're as fringe as Holocaust denial, even tiny increases in salience will be perceived, from the inside as significant. A jump from 0.01% to 0.02% is tiny, but still a doubling of interest.
Are we anywhere near the point where someone who isn't a conspiracy theorist or historical obsessive asks questions about the Holocaust? No. You mention a claim about the Nazis making soap out of victims - I've never heard of the idea that Nazis made soap from the bodies of murdered Jews, and I am, by normie standards, a WWII history nerd. (Simple test: I know what the Wannsee conference was. Most people do not.) I do not think that anyone near to what we might reasonably call the mainstream has heard of or cares about whether or not the bodies of Jews were turned into soap. As such, even if that's something widely believed and if there's been a change of mainstream academic opinion on it, I don't think it tells us anything about whether or not Holocaust denial is going mainstream.
When you say it's "understanding" and "thinking strategically", what you really mean it that it's generating plausible-looking text that, in the small, resembles human reasoning.
This is something I don't understand. The LLM generates text that goes in the 'thinking' box, which purports to explain its 'thought' process. Why does anybody take that as actually granting insight into anything? Isn't that just the LLM doing the same thing the LLM does all the time by default, i.e. make up text to fill a prompt? Surely it's just as much meaningless gobbledygook as all text an LLM produces? I would expect that box to faithfully explain what's actually going on in the model just as much as an LLM is able to faithfully describe the outside world, i.e., not at all.
Is this the turning point for WW2 revisionism entering the mainstream?
If 'WWII revisionism' means the idea that the Nazis weren't all that bad, or even Holocaust denial, then no. No, it is not.
I think it would help to avoid woolly euphemisms like 'WWII revisionism' and clearly state the thesis that is being considered. I do not think the public consensus that Nazi Germany was bad, that it committed hideous atrocities, and that it was right to destroy it is likely to change.
I'm not sure that shifting from regional to family-based conflict makes the distinction you want here? Ethnic or group identity can be tied to family or clan just as easily as it can to region. In fact it's usually a nexus - families or lines of ancestry, habitation in particular regions, shared language, shared cultural practices, and so on. What makes an ethnicity an ethnicity is a kind of overlapping Venn diagram.
Further I'm happy to say also that ethnicities can be fractal - contained within each other, going down level by level. 'Slavic', 'Germanic', or 'Turkic' are 'large' ethnicities, but they all contain within them smaller ethnicities. 'German' is a subset of 'Germanic', and then of course 'German' contains within it even smaller ethnic tribes. All Bavarians are German, all Germans are Germanic, and so on.
Likewise in Japan. All Satsumans are Yamato, but am I willing to say that Satsuman is a different ethnicity to Choshuans? It feels like asking whether Bavarians are a different ethnicity to Prussians or Saxons. It's largely in how you define 'ethnicity', and you can move that around however you like and I can't really argue.
What I would argue, though, is that regardless of whether we think the word 'ethnicity' is appropriate or not, historically Japan has been often divided, and people from different parts of Japan understood themselves to be meaningfully different to one another - certainly to the point of fiercely conflicting with one another.
I guess the point I would emphasise is that the idea of Japan being 'monoethnic' had to be produced, in large part by the Japanese themselves, through a struggle that took at least centuries. Japan, like every modernising nation, went through a process whereby national identity had to be constructed, often in the form of top-down policies of homogenisation and assimilation from the metropole.
It makes a nation like Japan seem somewhat boring, since it was mostly one ethnicity just interacting with itself instead of duking it out with dozens or hundreds of other ethnicities in a massive historical cultural bloodbath.
I'm going to nitpick that I don't think this is a historically accurate picture of Japan.
Firstly, Japan has historically possessed non-Yamato ethnic groups which were driven out, destroyed, or assimilated by the Yamato. The Ainu are the most famous one still extant, but historically you might consider the Emishi, the supposed 'shrimp barbarians' of northern Honshu who appear to have been either wiped out or assimilated into the Yamato whole.
Secondly, we should not assume that the Yamato specifically constitute a single 'ethnicity' who historically understood themselves to be a single united people. Modern Japan is the product of several generations-long processes of nation-building - the Tokugawa shogunate wanted to tamp down on regional identities and clan loyalties, and promoted a pan-Japanese identity, and the imperial government after Meiji took over intensified that process even further. Even today the process is not entirely complete or successful, and you will find very strong regional rivalries, including different languages and customs.
Go back a few centuries and I suspect that if you asked a group of people from northern Honshu whether they are 'one ethnicity' or 'one people' with a bunch of people from Kansai, and then you asked both groups whether they are 'one ethnicity' with a group of people from southern Kyushu, they would all say no. They might realise that they are more similar to each other than any of them are to Koreans, or to Han Chinese, or heaven forbid the Nanban, but I don't think they would see themselves as one homogenous group.
And just in general - I'm not sure how a person could look at the history of Japan and say that it's striking for lacking a "massive historical cultural bloodbath". Surely the history of Japan is a history full of distinct regional groups slaughtering each other? It feels like a form of special pleading to say "oh, that's different clans within a single cultural/ethnic group", rather than recognising that as different cultural groups. Perhaps they were all Japanese, in the same way that, say, the different states of the Holy Roman Empire were all German, but there is definitely a history there of rivalry, warfare, and bloodshed between groups that spoke differently, behaved differently, and felt themselves to be meaningfully different to one another.
I suppose I grant that there's a relatively 'low' use of the word intelligence that applies to things like the Doom AI. When I talk about the AI in a video game, I'm calling it 'intelligence', and in a sense I mean it.
But I don't see that as intelligence in a 'high' sense, if that makes sense? When I play a video game, the monsters are in some sense 'intelligent', but they are not intelligent in the sense that, for example, would be associated with possessing rights.
Maybe we should distinguish our terminology somewhat - perhaps the Doom demons are intelligent but not sapient? And I previously used the word 'intelligence' synonymously with sapience, rather than with this lower level of simulated agency?
In any case, I think I broadly agree with your conclusion. I think it is extraordinarily unlikely that in the near future we will produce artificial beings that can reasonably be said to be conscious, sapient, or an equivalent. I don't think we're going to produce any artificial people. But we are going to produce, and right now are in fact producing, machines that are capable of complex behaviour, and that these machines offer both potential and risk.
Well, I'm not sure that's how I would define 'intelligence'. I don't think that the laptop I am currently writing this message on is intelligent, even though it is capable of calculations that I am not.
But I certainly grant that a non-conscious machine could be very dangerous. That just doesn't strike me as the interesting part of AI hypotheticals.
You're perfectly free to not care about consciousness per se, but you were objecting to the hypothetical of someone objecting that a machine isn't conscious even as it kills them. But it's not clear to me why the hypothetical objector isn't correct. Consciousness clearly matters to that person, and the machine's lethality in no way indicates that it's conscious.
I am fully prepared to grant that a non-conscious machine that does not possess internal experience could be extremely complex and capable of destroying humanity.
I'm not sure how likely I think such a machine is, and I am skeptical that any such machine is likely to be produced in the next century, but it is philosophically conceivable. I just, to turn your argument back at you, don't really care about the hypothetical non-conscious supercomputer. It does not tell me anything about the thing I care about, which is personhood. So what if a supercomputer could conceivably destroy humanity? I've been able to imagine that for my entire life. Skynet is not a novel proposition. So... I just don't find the hypothetical genocide machine to be philosophically interesting.
To the extent that there's a controversy here, I suppose it's the practical question. "Would Skynet be conscious?" is a question you don't find interesting. "Would Skynet be capable of killing everyone?" is a question I don't find interesting. But "are we going to build Skynet?" is perhaps more relevant. As I said, I'm skeptical, and presumably you're more... I feel like 'optimistic' is the wrong word here, but you would estimate greater likelihood of that happening, I suppose?
Jesus Christ, some people won't see the Singularity coming until they're being turned into a paperclip.
Nuh uh, this machine lacks human internal monologues and evidence of qualia, you insist, as it harvests the iron atoms from your blood.
I find this argument strange, because being able to kill me is not evidence of a machine being conscious or intelligent. I could go and get myself killed by an LAW today, and if you asked me as I bled out and died, my body torn in two by an autonomously-fired rocket, I would still insist that the machine that killed me is not a person and does not possess internal experience. And I would be correct.
Whatever qualia are or are not, whether you think they're important or not, the question of qualia cannot be resolved or made irrelevant by a machine killing people. I should have thought that's obvious.
And to upscale from that a bit, I find it entirely imaginable that someone or other might invent autonomous, self-directed, self-replicating machines with no conscious experience, but which nonetheless outcompete and destroy all conscious beings. I can imagine a nightmare universe which contains no agents to experience anything, only artificial pseudo-agents that have long since destroyed all conscious agents.
There are already some novels with that premise, right? It doesn't use robots specifically, but isn't that the premise of Blindsight - that perhaps consciousness is evolutionarily maladaptive, and the universe will be inherited by beings without internal experience?
Thus I'm going to give the chad "yes". Maybe one day I get killed by a robot, and maybe that robot is not conscious and has no self-awareness. That it killed me proves nothing.
This is broadly correct, yes. Depending on how you count Aboriginals are somewhere between 3 and 4% of the total population, but "how you count" is contentious because generally it's just self-declared, and because Aboriginals get affirmative action in some places, if you have any Aboriginal descent at all you're probably going to tick the box.
So what 'Aboriginal' means depends a lot on the region - there are remote communities that are almost entirely of Aboriginal descent, but Aboriginals you find in the larger cities are likely to be mixed race. There's a media tic you may notice where they will describe Aboriginals as "a proud, X, Y, and Z man/woman", listing all the Aboriginal cultures that person identifies with, and obviously it's just a sliver for each one. Here's one example - "I'm a proud Bardi, Miriwoong, Yamatji, Noongar man". This is the equivalent of me saying, "I'm a proud English, Irish, Scots, French, German, and Swedish man", on the basis that I have some sliver of descent from each of them.
Anyway, because the total number of Aboriginal people is relatively small, and because a large proportion of them are heavily mixed-race, I doubt that, even supposing that primarily-Aboriginal-descended people have 10 or 20 less IQ than others, they skew the figures that much. As far as I'm aware there is an IQ gap, and you may speculate as to the causes as much as you like, but I just don't think there are enough of them for it to be a major factor.
I would argue, actually, that Revolutionary France developed extremely impressive state capacity, and it was its near-unprecedented level of mobilisation of resources that allowed it to fight off half of Europe. That wasn't something that automatically appeared in the absence of the king, but rather had everything to do with the systems of government of the new republic, some inherited from before the revolution, but some built anew as well. That level of organisation just doesn't come from nowhere.
More options
Context Copy link