4bpp
Now I am become a Helpful, Honest and Harmless Assistant, the destroyer of jobs
<3
User ID: 355
Have you heard about Poe's Law? Sarcasm doesn't work when it's equally plausible that you would say the same thing in earnest, and attaching a clichéd epithet like that without any deeper reasoning behind it is the sort of thing an LLM would do in earnest.
Anyway, unapologetic? Not at all. I just think it's a tradeoff I'm willing to make, even if it annoys you.
All's well. You're willing to annoy me with AI content, and I'm willing to annoy you by derailing your threads with speculations about AI content :)
The parent poster literally said "AI became the best mathematician alive". I was therefore not talking about my standard "for a successful mathematician". My standard for being the best mathematician alive does involve being as good as any other mathematician alive, which does in particular mean being >=Terry.
when the designated Reasonable Skeptic of the rationalist diaspora
Who? Aaronson? The "took his family to Disneyland and aborted the trip after having a meltdown over seeing people not wearing masks" guy? He's many things, but nobody is designating him any combination of "reasonable" and "skeptic".
(By the way, concurring with @2rafa's suspicion below, after you have acquired your reputation as an unapologetic meat proxy, it does feel extra suspicious when you make such catchy irresistible logit-attracting utterances whose substance seems bizarre when you actually think about it.)
Nah. If you were human and that were your only achievement, then sure, the attention-deficient celebrity chasing normie press might lionise you as such, but those in the know would not rank you above people who have demonstrated a lifetime of consistent progress on hard problems and good intuitions (such as Terry himself, and several of his associates).
The question really is if it's a can of worms you want to open. Both the main US tribes have some group of well-integrated foreigners that they are cozy with and derive some benefit from, and which has a horse in some conflict the US is tangentially involved in; so far neither has turned this into a political playing ball, but you could easily see a tit-for-tat start where Mamdanicore Democrats and mainstream Republicans deport people over having the wrong stance on Israel/Palestine, which in the end would just leave everyone worse off.
Isn't it perverse, to begin with, if businesses that rely on cheap (or any) parking cluster where the land is valuable? If their customers drive to them, why can't you have them in whatever location in the customer's driving radius has the cheapest land instead? Most European cities, and many American ones, already work like that, with shops that cater to driving customers setting up on marginal land on the city's periphery (e.g. nestled between industrial parks) instead of downtown.
Beyond what @Goodguy said below, the "building the torment nexus while railing against it" thing also seems at risk of being an instance of what they seem to call the "Goomba fallacy" these days (dumb name that tells you nothing about what it is, but that's all we have). Most EAs are not in a position to contribute anything of substance to the ongoing construction of the Torment Nexus, and might even be net delaying it by sapping funds that might otherwise go towards it.
I'm not sure if this is what you are referring to as "the initial offense," but magicalkittycat was banned for it. What is "jannie cover?"
It looks like he was banned for attacking @stoatherd, but the offense I am talking about is in the opening post of this thread (invoking a principle (apologise for false accusations that you make) that you provably do not consider to apply to yourself). "Jannie cover" is moderators banning people for calling this out.
I think you could analyse all that transpired in this thread as an extension of an initial offense (magicalkittycat's (mistargeted?) attack on stoatherd), but in that case, looking at the sequencing of posts, this reveals a more general structural problem with moderation. As far as I can tell, the chronology was:
-
magicalkittycat attacks stoatherd, seemingly based on a mixup of usernames, and they have the exchange in the thread you linked where the mixup comes out but magicalkittycat does not apologise
-
magicalkittycat makes a new top-level post attacking some outgroup senator for not apologising after making a false accusation
-
stoatherd calls him out on this
-
you ban stoatherd for his callout
-
I see that snapshot of the situation and start writing my response
-
you go back to the above thread and ban magicalkittycat for the original attack
-
I post my response to your ban of stoatherd
Between the beginning and the end of the chain, almost a whole day had passed. Not only is this a whole day in which the situation was allowed to fester in ambiguity (with plenty of opportunity for readers to build the expectation that this is yet another of the numerous cases where the moderators will not punish someone who manifestly deserved punishment), but the order in which you then proceeded to respond to the situation (punish the downstream reactions first, and only then move up the chain) created a window in which the above expectation would be confirmed (initial act unpunished, reactions cracked down on).
Moreover, due to how slow the response was, by the time the punishment for magicalkittycat's original post was dispensed, the community's attention window has long moved on, and hardly anyone will even see that ban. Therefore, it neither conveys deterrence to others who might commit a similar offense, nor a sense that justice was served to those who were offended by it (unless they were so obsessed as to save magicalkittycat's original post and make a note to come back to it the next day). The bottom line is broken window upon broken window was lined up and left a large number of readers with the diffuse sense that moderation is unfair.
Concrete suggestions to improve this situation (any subset of them would be an improvement):
-
when untangling heated situations where you are going to punish multiple people who attacked each other, moderate in chronological order
-
if a post was reported but the mod response is delayed because of internal discussion or nobody having time, have some way to flag "we received reports and are currently deciding how to proceed" to distinguish this state from "we consider this post good and won't do anything about it"
-
if moderation was slow to respond, consider the delay a mitigating circumstance for "vigilante justice" posts that offended back against the initial offender
-
link to related moderator action, so it doesn't wrongly appear one-sided if one side's punishment is prominent while the other's is buried (i.e. your modhat post against stoatherd should have linked to the modhat post against magicalkittycat)
I argued against this sort of ban when it was applied to someone calling out Dean, and I will argue against it here as well. I don't think prohibiting this kind of response is actually improving discussion quality. There are few things as infuriating as hypocrisy, where people seek to reap the discourse benefits of having principles when it is convenient, but refuse to pay the tax of being bound by them. The "heat" was already introduced when OP decided to be a hypocrite; stopping people from venting it will not let you cheat the Second Law of Thermodynamics, but just make them bottle it up and release it gradually over the course of future interactions with him (amplified by the additional anger over him getting away with the initial offense thanks to jannie cover). Presumably, this will result in OP feeling treated unfairly/diffusely persecuted by the community, which will in turn encourage him to needle his opposition more.
A sitting senator falsely accusing someone may be a bigger deal, but so is a sitting senator publicly admitting wrongdoing and apologising. Either way, the bottom line is that no amount of sophistry can really wash away the bad look of making ethical demands of someone else but showing no desire to be bound by them yourself at whatever smaller scale is applicable to you.
(On the meta level, a teaching moment. Even after conceding that you made a mistake, and even after this callout, you clearly are deeply averse to just saying sorry, even though it would resolve this altercation and improve your reputation considerably. Apologising to adversaries who you disdain and believe have wronged you in countless ways before is psychologically hard.)
I concede this wording was a bit hyperbolic/imprecise, but given his pivotal role in the Horthy abduction and Mussolini's rescue it's hard to imagine that someone would have been considered a higher-pedigree spy in the Hitler administration, at least.
I don't see as strong an effect as you seem to do - if you open German newspaper comment sections now, there are still people trying to inject conspiracy theories about how the Nord Stream bombing was really a Russian false flag, and they are not falling behind your median Mossad tale in complexity. Whatever difference there still is, in terms of the complexity hostile conspiracy theories impute to Israeli plots, can surely be explained by the Israelis and their cheerleaders themselves cultivating an image of superhuman competence and non-pickiness of means. Surely the idea that they recruited Hitler's former right hand man to assassinate Germans working for their adversaries would have been in the class of conspiracy theories you would call out the ridiculousness of, had they not publicised it themselves.
why would a nation capable of and intent on committing genocide wait until they were attacked to start doing so? Does that pattern-match to any historical genocide?
Because they require continued Western backing, if not for the genocide itself then at least to sustain their long-term economy and strategy. I can't think of any historical precedent for the general configuration that necessitates this strategy, where an overwhelming global hegemon's actions are constrained by a somewhat well-informed population's perceptions of morality - the window perhaps opened with Vietnam, and is now closing as tribalism and consensus-manufacturing tech are catching up. Any candidate examples would therefore have to be putative genocides, by nations the US could be expected to intervene for or against, in the last 50 years. That's not a big pool, and what there is is largely still hot culture war material that it's impossible to analyse dispassionately.
I was not particularly thinking of actual Israelis, Ukrainians or Russians when making that list. All of those deflections could be sampled from a smattering of Western European and American gentiles, in defense of a team they chose to support. (Was @FtttG himself even Jewish, on that matter?) You could add Chinese/Uyghurs to the list as well, except that topic area has largely dropped from the public consciousness.
Is this not just the general pattern of how everyone handles instances of violence (or, more generally, violations of something they claim to be a principle they hold when they actually don't) against an enemy? Pro-Israelis do the same thing about "Israeli military sniping women and children for sport" reports; pro-Ukrainians pulled it about North Stream, Ukrainian assassinations of their own opposition abroad, and whenever their own attacks hit civilians; pro-Russians could be seen pulling it in real time about the invasion of Ukraine itself, among others; and so on and so forth.
We are all by and large beasts wishing for peace and prosperity for our friends and misery and death for our enemies, and the pretense that there is any ethics or principles beyond that only serves to make us beastlier still, like the trope of the "why you hittin' yourself?" bully.
(On the other hand, of you are willing to forego treating your arguments, empathy and every other act of communication as soldiers, there isn't even any particular contradiction between being happy for the passengers and thinking Israel is in the wrong! You can cheer on both the Russian and the Ukrainian soldier cleverly dodging a drone, even though each act is sure to set back the cause of the respective other by a little and threading the needle to support both causes at once is hard. These are fellow humans; they fought to survive against stacked odds; by luck, quick thinking, skill and coordination, they managed to pull it off.)
It's not like he gave any examples of wars between Communist countries. The most obvious examples I am aware of all involved the China-Vietnam-Cambodia triangle, which I did cover with examples; I then added some more obviously nationalist communist countries because otherwise there would not be enough content.
Many communist countries were not actually particularly globalist (North Korea in particular, but also e.g. Albania; North Vietnam came out of a nationalist movement of colonial liberation that was just desperate for backers after the US sacrificed them for the sake of their European postwar order political project; I think Laos might even still forbid its citizens from marrying foreigners). On the other hand, what are actually your Western examples? I won't take mere decades as evidence of durability here, on historical clocks it should at least be 100 years to transcend the realm of suspected expedience. England and the US are, well, the same nation; I wouldn't consider German-Austrian alignment to be evidence that nationalists can trust other nations either. Proper nationalism has not even been around for that long (it is commonly taught as really taking off with the revolutions of 1848, though arguably the English and French both woke up to national consciousness earlier than that), and the whole story of Europe is that up until the Pax Americana the alliances were constantly shifting.
Would European Red-analogues, if they were had their day in the limelight, even be natural allies to American Reds? The existence of a "Blue Ummah" is, in a way, the intrinsic advantage of Blues; globalists everywhere have common cause, but nationalists can only have temporary allies of convenience. I forget if it was SSC or some other blog in the general sphere that proposed the thought experiment of some lizard species where a subset could be identified as cooperate-bots with a visual signal only visible to fellow cooperate-bots and thereby outcompeted all the others (as a metaphor for certain mind-killing minorities), but it's kind of like that except the signal is not even secret.
Because the effort that is being fairly matched is cognitive effort, not incidental physical effort.
I don't deny text has more uses; if the point is to actually obtain information I seek, I don't care for the provenance (that's why I'm happy with a lot of AI-generated documentation, and will ask Gemini to sift through search results). However, the main point of reading and writing on a forum like this, or blog posts on SSC, is not seeking and providing specific information.
You are ignoring the widely repeated argument against it that is based on the attention economy - "I read it because you took the time to write it" etc. - which I find much more compelling. Adjacent sentiments were reasonable long before genAI, which I find to disprove the "psychosis" angle: if my long-distance girlfriend wanted to call me to give me an identical 10-second good morning greeting every day, that would be much easier to accept than if she made a 10-second recording and insisted that I play it back at a fixed time every day, and if a friend wrote me a 10-page rant on some topic, I would feel much more compelled to read it than if be sent me a 10-page blog post by someone else that he thinks happens to summarise his exact thoughts on a subject.
More laconically, surely you can't complain if I answer your "I asked AI to write this post for me" by "I asked AI to read your post for me".
I would put the suburban threshold at 2 rides/hour rather than 4. Hell, at the height of the train maintenance crisis a few years ago, some inner-ring parts of Berlin were reduced to 1/20min at peak hours, and I've lived in areas with 20- and 30-minute period transit that were nevertheless within 20 minutes of the centre and largely housed urban commuters.
Did the part of Germany you lived in actually have good private schools? In the East, the best public schools are squarely superior, as evidenced by the intake of math/science olympiads, and private schools are mostly considered a way for rich people and politicians to sequester their kids.
The whole point is that you needn't worry that you get lynched if you are protected by a tireless AI-controlled array of drones and Boston Dynamics dogs. To expect that the rabble will find a way around it is to believe, against all odds, that humans will remain superior in some way after all.
- Prev
- Next

I do wonder how the reaction will be when, almost inevitably, some PhD student or postdoc gets scooped on a problem they had gone all in on and commits suicide. It's not like OpenAI seems to be following the standard process of extending feelers through the field to get in touch with any existing noosphere-homesteaders before solving these.
More options
Context Copy link