site banner
Advanced search parameters (with examples): "author:quadnarca", "domain:reddit.com", "over18:true"

Showing 25 of 332513 results for

domain:savenshine.com

We are looking at this from two different angles. My angle helps people. Your angle, which seems to prioritize protecting the LLM from the 'insult' of a simple metaphor, actively harms user adoption. My goal in using the parrot model is to solve a specific and very common point of frustration - the anthropomorphising of a tool. I know the parrot shortcut works, I have watched it work and I have been thanked for it.

The issue is that humans - especially older humans - have been using conversation - a LUI - in a very particular way their entire lives. They have conversations with other humans who are grounded in objective reality, who have emotions and memories, and therefore when they use a LUI to interact with a machine, they subconsciously pattern match the machine to other humans and expect it to work the same way - and when it doesn't they get frustrated.

The parrot model on the other hand, tells the user 'Warning: This looks like the UI you have been using your whole life, but it is fundamentally different. Do not assume understanding. Do not assume intention. Your input must be explicit and pattern-oriented to get a predictable output.' The parrot doesn't get anything. It has no intentions in the sense the person is thinking of. It can't be lazy. The frustration dissolves and is replaced by a practical problem solving mindset. Meanwhile the fallible intern exacerbates the very problem I am trying to solve by reinforcing the identification of the LLM as a conscious being.

The beauty is, once they get over that, once they no longer have to use the parrot model to think of it as a tool, they start experimenting with it in ways they wouldn't have before. They feel much more comfortable treating it like a conversation partner they can manipulate through the tech. Ironically they feel more comfortable joking about it being alive and noticing the ways it is like and unlike a person. They get more interested in learning how it actually works, because they aren't shackled by the deeply ingrained grooves of social etiquette.

You're right that metaphors should be analyzed for fitness, but that analysis requires engaging with the metaphor's intended purpose, not just attacking its accuracy literally. A metaphor only needs to illuminate one key trait to be effective, but the parrot goes a lot further than that. It is in fact fantastic at explaining the spiky profile of LLMs. It explains why an LLM can 'parrot' highly structured Python from its training data but write insipid poetry that lacks the qualia of human experience. Likewise I could train a parrot to recite 10 PRINT "BALLS"; 20 GOTO 10, but it could never invent a limerick. It explains why it can synthesize text (a complex pattern matching task) but can't count letters in a word (a character level task it's not trained to understand). Your analysis ignores this context, seemingly because the metaphor is offensive to an aspirational view of AI. But you're attacking a subway map for not being a satellite image. The resolution is drastically reduced yes - this is a selling point, not a flaw. Cultural cachet drastically outweighs accuracy when it comes to a metaphor's usefulness in real world applications.

And do you want to know another animal with a clearly non human form of cognition? A parrot. How did you skip over crows and dolphins to get to octupi, animals with an intelligence that is explicitly not language based, when we are talking about language models? Unlike an octopus, a parrot's intelligence is startlingly relevant here (my mentioning of parroting was just an example of how a parrot has been used as a metaphor for a non-thinking (or if you prefer, non-feeling) pattern matcher in the past.) Using a LUI a parrot can learn complex vocalisation. They can learn mimicry and memorisation. They can learn to associate words with objects and concepts (like colours and zero). They can perform problem solving tasks through dialogue. Is it just because octupus intelligence is cool and weird? Because that just brings me back to the difference between evangelising llms and helping people. You want to talk up llms, I want to increase their adoption.

Shaming users for not having the correct mental model is precisely how we end up with people who are afraid of their tools - the boomers who work out calculations on a pocket calculator before typing them into Excel, or who type 'Gmail login' into the Google search bar every single day. As social media amply demonstrates, technical accuracy does not aid in adoption, it is a barrier to it. We can dislike that from a nerd standpoint, which is why I admired your point in my original post (technically correct is the best kind of correct!) but user adoption will do a lot more for advancing the tech.

I would argue that the Trump administration is not abusing the statute, at least in the case of Khalil. The statute specifically allows for deporting people for protected speech, and Mahmoud Khalil -- a "student" who seemingly did not come to the US to study but rather to stir up trouble on campus for the purpose of affecting US foreign policy -- is the exact sort of person it was meant for.

I'd be fine for the statute to be declared unconstitutional on its face. I suspect it will not be, however; the First Amendment protects citizens and probably permanent residents, but possibly not those on temporary visas.

If that Libertarian world were less than completely theoretical, or if the Libertarian movement gave any indication that they had a method of getting us there, this might weigh heavier of the deliberations of former libertarians.

(5) LEWD OR LASCIVIOUS MOLESTATION.--

(a) A person who intentionally touches in a lewd or lascivious manner the breasts, genitals, genital area, or buttocks, or the clothing covering them, of a person less than 16 years of age, or forces or entices a person under 16 years of age to so touch the perpetrator, commits lewd or lascivious molestation.

An offender 18 years of age or older who commits lewd or lascivious molestation against a victim 12 years of age or older but less than 16 years of age commits a felony of the second degree, punishable as provided in s. 775.082, s. 775.083, or s. 775.084

See also

(7) LEWD OR LASCIVIOUS EXHIBITION.--

(a) A person who:

  1. Intentionally masturbates;
  2. Intentionally exposes the genitals in a lewd or lascivious manner; or
  3. Intentionally commits any other sexual act that does not involve actual physical or sexual contact with the victim, including, but not limited to, sadomasochistic abuse, sexual bestiality, or the simulation of any act involving sexual activity in the presence of a victim who is less than 16 years of age, commits lewd or lascivious exhibition.

(c) An offender 18 years of age or older who commits a lewd or lascivious exhibition commits a felony of the second degree, punishable as provided in s. 775.082, s. 775.083, or s. 775.084

I could go on with all the Florida statutes he could have been charged under, but I don't have all night.

Sûre, I eat hunted meat when practical- and happily point this out to ethical vegans- for health and animal cruelty reasons. I have western-typical ideas about eg dogfighting even if I think no-kill shelters are impractical and cause too many problems.

How much of what was admitted to was actually illegal at the time? It’s not illegal to show people your underwear, even if it will likely result in being asked to leave a public place- but had these girls went to Walmart like that they would have been charged with trespass, not a sexual performance or whatever. Masturbating in front of the girls and touching them with a vibrator might legitimately have been loopholes in the law- it’s entirely possible that the girl would have had to have touched his penis to trigger statutory.

On an intellectual level? I'm well aware that people are using LLMs for smut. I just never looked into the... specifics of it. I regret to say that I think the I read all of the Evangelion stuff, I got off the train when the beastiality station was the next stop. Or is it even beastiality if it's a dragon fucking a pony? Only Satan knows.

If I want titillation, I'm opening PornHub. Text isn't my jam, but if it's yours then I'm not one to actually judge. From a purely technical standpoint, it all came across as competent, probably way better than the average literotica writer.

If citizenship is not a matter of ethnic belonging (i.e. an intersection of familial and cultural ties), it must then be ideological, which necessarily means exercising discretion and control over the ideologies of people allowed in

This is a false dichotomy. There are a lot of other possibilities around which to organize citizenship. In the Roman republic and empire, it was for example build around collective military aid: polities on the Italian peninsula which were subjugated by Rome and fought side by side with the legions were eventually granted citizenship. In the French foreign legion, it is individual military aid: mercenaries who are willing to die for French interests get French passports for their service. Or a lot of cultures were willing to integrate women who married their citizens, willingly or otherwise. In Tír na nÓg , citizens and immigrants share a metatype. In Israel, immigrants and citizens (mostly) share a religion (which is only an ideology in the widest sense of the word).

Of course, if America has an ideology, it is the ideology of the American dream. The idea that an immigrant who arrives with little more than the clothes they are wearing can through hard work thrive in the land of capitalism and freedom. Having opinions about which group should murder which somewhere in the old world seems pretty orthogonal to that.

Some rights are clearly citizen rights -- the right to vote is a clear example. Some other rights are basically human rights, and should naturally apply to any human under the power of the state. The right to free speech, freedom of religion, due process, confront their witnesses, or the prohibition of cruel and unusual punishments are examples of the latter.

Ideally, I would not want the government to change people's immigration status based on any actions which are not actually criminal (or at least narrowly concerning immigration, such as "lying on your visa application"). A world in which each free country has an LLM which searches if visa applicants have made any statement the ruling party does not like and automatically rejects them is not a freer world than one where no free country does such a thing.

I think this is the crux of the disagreement. We both have axiomatically different reasons for holding our positions. There’s no reasoning across that gap. I’m not going to convince you, and you aren’t going to convince me.

However, it seems from the last paragraph that if we could provide adequate amounts of meat/protein with less suffering you would be in support of that. As would I think most users on this form. That seems like a much more fruitful focus of vegan political efforts than the bullshit that PETA wastes its funding on.

So you'll often see libertarians defending Corporations, Universities or NGOs for trampling your rights (It's a private entity, it can do whatever it wants!), while they condemn the government for doing the same

To play devil’s advocate(and to be clear I am not a libertarian, even if I don’t like the government very much)- a lot of the reason corporations and universities are on one side of the culture war in particular is due to government policy making it hazardous to take the other side. Theoretically in a libertarian world conservative universities wouldn’t be limited to small religious liberal arts colleges.

I do not think animal suffering is a good, or even neutral thing, exactly, but I don’t think it’s in the same category of human suffering- it’s an ordinary evil similar to soil degradation from mono cropping or similar. No, I don’t see gorillas or elephants or whales as meaningfully different in that regard, even if they’re as smart as a child. I don’t care if harambe or dumbo or shamu are as smart as a child they don’t have the moral worth of one, it’s a qualitative difference not quantitative. To my way of thinking elephants are rarer than rats and so killing one should be a much higher bar to clear, but there isn’t a moral problem with shooting a depredating elephant from a helicopter in the same way that there isn’t a moral problem with setting a rat trap.

That’s why I don’t worry about shrimp eyestalk ablation or farrow crates. Doing these things for no reason would be profoundly evil. Doing them to have affordable animal protein is as necessary as tilling fields; and having almost everyone in society eat meat every day is a straightforwardly good thing so much so as to be miraculous, for child development reasons if nothing else.

how do we address the issue I mentioned of idealogy changing constantly?

This is the price we pay for being a propositional nation; the proposition changes as the populace does...when the melting pot itself melts and Jeffersonian democracy wears away to nothing, there's a fight over how to replace it. And like most fights, there's swings back and forth until a new consensus emerges.

I don't think there is a complex allegory hiding behind the Avatar movies.

that would imply it being intentional

I see rather it as hilarious exploration of completely unintentional rejection of nature and accidentally revealing that self-proclaimed environmentalist does not understand nature, rejects it and prefers advanced tech.

The gourmet vegan diet you’re referring to is available to most regular vegans, though. Rice and beans are, famously, among the cheapest things in the grocery store, veggies aren’t expensive, and even the expensive parts like avocados are still cheaper than meat.

Regular vegans don’t eat like gourmet vegans, I suspect, for the same reasons regular meat eaters don't eat like my home-cooked French-Mexican fusion food made with wild game.

I'm about 75% of the way through wiring TRON helmet bike LEDs and the second joint I soldered together is already kind of janky and the connection now winks in and out as I wiggle the LED strips. Either my first few joints just sucked but it'll get better as I finish or I need to give up on this approach and use those bulky connectors kits after all.

Previous discussion of my possibly-too-jank method so far here https://www.themotte.org/post/2271/tinker-tuesday-for-july-22-2025/349091?context=8#context

It's pretty good, as is the movie; Soviet Batman is awesome.

Ok, now look up “BNWO.”

While i agree with the overall thrust of your critique i want to harp on this bit

Will he change his mind if someone presents a chess-focused LLM with a high ELO score?

...i think that part of the problem is a wide-spread failure on the part of Freddie and the wider rationalist community to think clearly and rigourousl about what "intelligence" is supposed to mean or accomplish. It is trivially true that by restricting an LLM's training data to valid games of chess documented in the correct notation and restricting it's output to legal moves you can create an LLM that will play chess at a reasonably high level. Where as a LLM trained on an appreciable portion of the entire internet and few if any restrictions on its yout put will lose to a computerprogramwritten in the 70s. The issue is that your chess an llm is not going to be a general tool that can also produce watercolor paintings or summarize a YouTube videos its going to be a chess tool and must be evaluated within that context. If stockfish can reach a similar ELO using less compute why would you use anything else? One of the weird quirks of LLMs is that the more you increase the breadth of thier "knowledge"/training data the less competent they seem to become at specific tasks for a given amount of compute witch is the exact opposite of what we would expect from a thinking reasoning intelligence, and i think this points to a hole in the AI booster and AI doomers reasoning which is that the G in AGI is arguably far more operative than the I.

It's a curiosity because without principles, what makes someone choose any particular side to begin with?

Familial/tribal/ethnic loyalties? Nobody is born into a void, into the "view from nowhere"; we're all born into a particular place, a particular family, particular conditions; embedded in a specific social context, full of unchosen bonds and obligations, which indelibly shape who we are.

You (generic/rhetorical "you," not making any assumptions here) love your family not because they're "the best family" according to some prior metric, you love your family because they're yours. Much the same with patriotism. To quote Chesterton, "Men did not love Rome because she was great. She was great because they had loved her."

Wait, seriously? You are an LLM evangelist and you have never even thought about doing this? Come on. The internet is for porn; there must be something you want to read.

Meanwhile FIRE has been pretty consistent in criticizing both the left and right, and even defending their opponents right to speech. It's like the early ACLU protecting the rights of KKK.

This attitude seems certainly praiseworthy. The principled org willing to piss off people on both sides of the CW is much more credible than any organization which only cares about their pet issue when it benefits their side.

Also, calling a freedom of speech organization FIRE (in caps) is brilliant naming.

We could do a Mr Glass / Unbreakable cosplay lol

True story; in college me and my buddies were fucking around in the park at night and my friend jumped out of a tree and accidentally did a flying knee with his full weight right to my collarbone from 10ft up, hurt like shit but I got up and we kept messing around, went and got beers.

Next morning I went to class and then went to work. My shoulder was killing me but I just kept working.

Finally after a week without it getting better I went to the doctor after my girlfriend wouldn’t let up.

Turns out I had a hairline fracture and I had been just walking around like it was nothing. Only time I’ve ever “broken” something.

Both my girls are built like tanks too, so it’s definitely genetic.

But the above makes me wonder - the initial girl (Jane Doe) was 15, yes? So if there is evidence (or at least accusations) that he got 15 year old girls to strip down to their underwear, 'massage' him while he was naked, and he used vibrators on them and/or jerked off in their presence, then paid them - surely that is something more than "well he did a little bit naughty in paying for a massage from an unlicensed person"?

I'll preface this by saying that I don't have access to full transcripts of OPR interviews with the people from the State's Attorney's office, and while the grand jury transcripts have been released, I can't find anything specifying what charging options were presented. But my speculation based on what has been released is this: The State's Attorney was concerned about the ethical implications of charging Epstein with prostitution-related offenses without charging the prostitutes themselves. Krischer had previously charged girls as young as 14 with prostitution, but he clearly recognized that the girls here had been taken advantage of, and the office was uneasy about charging witnesses who came forward.

I think that more importantly, though, Epstein was already offered a misdemeanor plea he refused, and if he tried to nail his ass to the wall he'd be looking at a trial that would be a fucking mess. You mention credibility issues, but it's not just a matter of whether the jury believes the girls, but whether the jury believes they were abused. Remember, this was a time when the public was sneering at kids like this on a daily basis, as Maury Povich sending incorrigible teens off to boot camp was mainstream (if lowbrow) entertainment. It wasn't so much that a jury wouldn't believe what happened, but that they wouldn't be able to view the girls as victims. Adding to the problem, the case hinges on the girls testifying to all of this bad behavior in open court, and even if you can keep some of it out, the fact still remains that they have to admit to prostituting themselves, some on multiple occasions, and to recruiting other girls to do the same thing.

These days juries are much more sympathetic to the idea that kids in these kinds of situations often have serious problems, and it's easier to paint a guy like Epstein as someone who recognized how vulnerable they were and took advantage of them. But it wasn't clear yet in 2006. Federal prosecutions require a grand jury indictment, but in state court the normal procedure is to file an "information", which results in some kind of preliminary hearing in front of a judge to determine if there's probable cause to go to trial. Grand juries are only used in unusual situations; they can be investigative tools since witnesses can testify under subpoena, and they're often used for complicated cases involving organized crime, public corruption, etc. I think that the decision to take the Epstein case to the grand jury was a consequence of the State's Attorney's uncertainty about how a jury would react to the evidence, especially in the face of an aggressive defense. It would give them a chance to defer the charges to somebody else, rather than filing the charges police wanted them to file and taking the chance that the case would fall flat.

As I said, I don't know what charging options the grand jury was given, but for the sake of argument I'll assume that the charges the police were pushing for were among the options. After the transcripts were released last year, prosecutor Lanna Behlolovick was criticized extensively in the media for apparently sandbagging her case by only having two girls testify and bringing up all the bad behavior. I disagree with this assessment. I think she knew that the defense was going to bring it up at trial and she wanted to see how a jury would react. One difference between grand juries and trial juries is that grand jurors have the opportunity to question witnesses, and the questions asked by the grand jurors don't evidence much sympathy. Some of them made glib comments to that effect. This was especially the case when a detective presented the evidence of other girls who had been abused (hearsay is admissible in grand jury proceedings), and they weren't at any risk of offending the girls directly. If they were offered a full slate of charges but only indicted on the solicitation charge, it's evidence that the case was a loser. There's also evidence that the grand jury's unwillingness to indict factored into Acosta's decision to seek a pre-indictment plea, since a Federal jury wasn't likely to be any better on that front. Having immersed myself in this whole mess, it causes me to wonder what the public reaction would be now if Epstein had been charged with serious crimes but acquitted. Would this even be something we're talking about now?

Besides general American values, how do we address the issue I mentioned of idealogy changing constantly?

Evidence indicates that we fight, with ever-increasing viciousness, over unilateral enforcement of our tribally-preferred ideologies, and the devil take the hindmost.

They could have been a game changer, but that was conditioned on them containing child pornography, or worse, containing evidence that he transmitted CP over the internet.

That's such a huge game changer, and should be painfully obvious to any prosecutor with experience. Possession of CP triggers all kinds of mandatory minimums that increase the prosecution's leverage by absurd amounts (federally, anyway. I don't know if Florida law has mandatory minimums for it). Even a 1% chance of the computers containing it dramatically changes the case, and it's a tiny mental stretch made by prosecutors every day to say, "hey, we're investigating this guy for sex crimes, perhaps he's a CP collector, too."

You could be right that his homemade videos at the time would be worthless in strengthening the case against him with regards to the known victims. But the chance of finding CP has had every prosecutor I've dealt with jumping at the chance to seize every single electronic device possible from suspects. Proving a hands-on offense with a victim with credibility and reliability issues is tricky; proving possession of CP doesn't have those problems. I don't have training materials from USAO from that era, but even in 2005, I have a hard time believing it wasn't common knowledge that finding CP on a suspect's computer was the "easy win" button.