This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
I thought I explained it pretty well, but I will try again. It is a cognitive shortcut, a shorthand people can use when they are still modelling it like a 'fallible human' and expecting it to respond like a fallible human. Mode collapse and RLHF have nothing to do with it, because it isn't a server side issue, it is a user issue, the user is anthropomorphising a tool.
Yes, temperature and context windows (although I actually meant to say max tokens, good catch) don't come up in normal conversation, they mean nothing to a normie. When a normie is annoyed that chatgpt doesn't "get" them, the parrot model helps them pivot from "How do I make this understand me?" to "What kind of input does this tool need to give me the output I want?"
You can give them a bunch of additional explanations about mode collapse and max tokens that they won't understand (and they will just stop using it) or you can give them a simple concept that cuts through the anthropomorphising immediately so that when they are sitting at their computer getting frustrated at poor quality writing or feeling bad about ignoring the llms prodding to take the conversation in a direction they don't care about, they can think 'wait it's a stochastic parrot' and switch gears. It works.
A human fails at poetry because it has the mind, the memories and grounding in reality, but it lacks the skill to match the patterns we see as poetic. An LLM has the skill, but lacks the mind, memories and grounding in reality. What about the parrot framing triggers that understanding? Memetics I guess. We have been using parrots to describe non-thinking pattern matchers for centuries. Parroting a phrase goes back to the 18th century. "The parrot can speak, and yet is nothing more than a bird" is a phrase in the ancient Chinese Book of Rites.
Also I didn't address this earlier because I thought it was just amusing snark, but you appear to be serious about it. Yes, you are correct that a parrot can't code. Do you have a similar problem with the fact a computer virus can't be treated with medicine? Or that the cloud is actually a bunch of servers and can't be shifted by the wind? Or the fact that the world wide web wasn't spun by a world wide spider? Attacking a metaphor is not an argument.
I've explained why I think the parrot is a terrible metaphor above. And no, metaphors can vary greatly in how useful or pedagogical they are. Analyzing the fitness of a metaphor is a perfectly valid, and in this case essential, form of argument. Metaphors are not neutral decorations; they are cognitive tools that structure understanding and guide action.
A computer virus shares many properties with its biological counterpart, such as self-replication, transmission, damage to systems, the need for an "anti-virus". It is a good name, and nobody with a functional frontal lobe comes away thinking they need an N95 mask while browsing a porn site.
The idea of the Cloud at least conveys the message that the user doesn't have to worry about the geographical location of their data. Even so, the Cloud is just someone else's computer, and even AWS goes down on rare occasions. It is an okay metaphor.
The Parrot is awful. It offers no such explanatory power for the observed, spiky capability profile of LLMs. It does not explain why the model can write functional Python code (a task requiring logic and structure) but often produces insipid poetry (a task one might think is closer to mimicry). It does not explain why an LLM can synthesize a novel argument from disparate sources but fail to count the letters in a word. A user equipped only with the parrot model is left baffled by these outcomes. They have traded the mystery of a "fallible human" for the mystery of a "magical parrot".
I contend that as leaky generalizations go, the former is way better than the latter. An LLM has a cognitive or at least behavioral profile far closer to a human than it does to a parrot.
You brought up the analogy of "parroting" information, which I would assume involves simply reciting things back without understanding what they mean. That is not a good description of how the user can expect an LLM to behave.
On an object level, I strong disagree with your claims that LLMs don't "think" or don't have "minds". They clearly have a very non human form of cognition, but so does an octopus.
Laying that aside, from the perspective of an end-user, LLMs are better modeled as thinking minds.
The "fallible but knowledgeable intern" or "simulation engine" metaphor is superior not because it is more technically precise (though it is), but because it is more instrumentally useful. It correctly implies the user's optimal strategy: that performance is contingent on the quality of the instructions (prompting), the provided background materials (context), and a final review of the output (verification). This model correctly guides the user to iterate on their prompts, to provide examples, and to treat the output as a draft. The parrot model, in contrast, suggests the underlying process is fundamentally random mimicry, which offers no clear path to improvement besides "pull the lever again". It encourages users to conceptualize the LLM as a tool incapable of generalization, which is to ignore its single most important property. Replacing a user's anthropomorphism with a model that is descriptively false and predictively useless is not a pedagogical victory. It is swapping one error for another, and not even for a less severe one to boot.
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.
Touché. I walked into that one.
Look, come on. We are literally in a thread dedicated to avoiding Bulverism. Do you honestly think I'm out here defending the honor of a piece of software? My concern is not for the LLM's public image. Sam Altman is not sending me checks. I pay for ChatGPT Plus.
I think the charitable, and correct, framing is that we are both trying to help people use these things better. We just disagree on the best way to do that. My entire point is that the "stochastic parrot" model, while it might solve the one specific problem of a user getting frustrated, ultimately creates more confusion than it solves. It's a bad mental model, and I care about users having good mental models.
You're right that a metaphor is a subway map, not a satellite image. Its value is in its simplification. But for a subway map to be useful, it has to get the basic topology right. It has to show you which stations connect. The parrot map gets the topology fundamentally wrong.
It tells you the machine mimics, and that's it. It offers zero explanation for the weird, spiky capability profile. Why can this "parrot" debug Python but not write a good joke? Why can it synthesize three different academic papers into a novel summary but fail to count the letters in a word? The parrot model just leaves you with "I guess it's a magic parrot". It doesn't give the user any levers to pull. What's the advice? "Just keep feeding the parrot crackers and hope it says something different?"
Compare that to the "fallible but brilliant intern" model. It's also a simplification, but it's a much better map. It correctly predicts the spikiness. An intern can be a world-class expert on one topic and completely sloppy with basic arithmetic. That feels right. More importantly, it gives the user an immediate, actionable strategy. What do you do with a brilliant but fallible intern? You give them very clear instructions, you provide them with all the necessary background documents, and you always, always double-check their work for anything mission-critical. That maps perfectly onto prompt engineering, RAG, and verification. It empowers the user. The parrot model just leaves them shrugging.
I'm pretty sure I haven't done that. My frustration isn't with your average user. It's with people who really should know better using the term as a thought-terminating cliche to dismiss the whole enterprise.
If my own grandmother told me she was getting frustrated because "Mr. GPT" kept forgetting what she told it yesterday, I wouldn't lecture her on stateless architecture. I'd say something like, "Think of it as having the world's worst long-term memory. It's a total genius, but you have to re-introduce yourself and explain the whole situation from scratch every single time you talk to it."
That's also a simple, not-quite-accurate metaphor. But it's a better one. It's a better map. It addresses her actual problem and gives her a practical way to think that will get her better results next time. It helps her use the tool, which is the goal I think we both agree on.
I'm pretty sure you said people like me are less intelligent than a parrot and that you hope we get mauled by a tiger. You did not specify that it was only directed at those using it to dismiss using AI, it was anyone using the term unironically. If I felt shame like normal people I would have simply stopped doing it instead of defending it - and I would no longer be helping people stop anthropomorphising a tool.
You lay out your complex 'fallible intern' model as the superior model. It can debug code and synthesise academic papers, it has a mind, though unlike any we know. You say we need to teach people to give clear instructions, provide background documents, and verify all work. But when you imagine talking to your own grandmother - a perfect example of a novice user - what do you do? You drop the intern model completely in favour of a genius with the world's worst memory. Why?
Because you know the intern model is too complicated. You know it doesn't work for a normal person. You'd never actually saddle your grandmother with the mental load of dealing with an intern who is an amnesiac - and is also a compulsive liar who has mood swings, no common sense, and can't do math. You give her a simple tool for the problem. But your tool deals with the symptom, mine deals with the cause.
I believe that you are trying to help people too, but you really are prioritising defending your model first. It might work great with techbros or the techbro adjacent, but even you drop it when you imagine a real world situation with a novice.
And I have to say, if I told you I'm not biased towards Teslas, Elon doesn't send me cheques, and in fact I just paid money for one, how wide would your eyes go as you attempted to parse that?
Look, I think it's quite clear that my statement about tigers was hyperbole. You seem like a perfectly nice guy, while I wouldn't jump into the ring to save you, I'd throw rocks (at the tiger) and call for paramedics.
That is the nice thing about being able to compartmentalize one's combative online persona from being an actually nice and easy-going person in reality. There are very few people I would actually watch and let die, and they're closer to Stalin than they are to people I disagree with on a forum for underwater basket weaving.
If this were a perfectly realistic scenario, my conversation would go:
"What the fuck. Is that a ghost? I thought your ashes were somewhere in the Bay of Bengal by now."
Do you understand why I paraphrased what is usually a more nuanced context-dependent conversation IRL? If my granny was actually alive, I would probably teach her how to use the voice mode in her native language and let her chill.
Uh? I don't know. If you have a reputation for doing that, I genuinely do not recall. I am very active here, but I do not remember that without actually having to open your profile.
Noted. You won't take back either statement - I am still dumber than a parrot (given the retreat you have been on over the last few posts I guess that's score 1 for parrots?) and you still want to see me meet a tiger outside of its cage, but you would throw rocks at it.
I am familiar with hyperbole. I am also familiar with the mechanics of shaming. I think you are too and you know that isn't a defence. Shame often uses hyperbole to express the level of emotion of the shamer and to trigger a more visceral reaction in the shamed. Can I start ending my arguments on the motte with die in a fire if I promise it's rhetorical?
On the topic of your grandma, you have my condolences. Retreating to literalism is just more condescension though, it's not an argument I will engage with, particularly when I already noted the hypothetical nature of the exercise. I will simply point out that you had the opportunity to deploy your intern model in a hypothetical with a novice user and you refused - twice now.
You have not needed to argue any of this.You're clearly capable of nuance when you want to be - over the past day you've written however many words on MIAD in that other thread and also given me a detailed breakdown of how and why you'd throw rocks at a tiger. You chose, after explaining the superiority of the intern model, not to use it. After having the discrepancy pointed out, you chose again not to use it. You can't imagine using it because it does not work as a cognitive shortcut, case closed. High five Sam Waterston. Created by Dick Wolf.
Lastly, my point about Tesla is that the fact that you are willing to pay for ChatGPT plus is a mad defence against the claim that you are evangelising on its behalf. You don't need to pay someone to advertise your product if they are already paying you, that's advertising 101 - you let the principles of brand fusion and post purchase rationalisation do their thing, eventually reinforced by the sunk cost fallacy. As these things go it's closer to a confession than it is to a defence.
Dude. Do you lack a sense of humor? This isn't intended to be an insult, but I am genuinely confused. I clarified that quite a bit of my apparent hostility towards bird fanciers is a joke. I can get annoyed by certain types of people at times, but I don't wish death upon them. I also make it a general policy not to fistfight tigers on behalf of strangers, you've gotta be close family or a loved one to make me consider that. I do not think we're family, and I do not think I sleep with you.
You're taking this to a place of literalism that's honestly baffling. "Can I start ending my arguments on the motte with die in a fire if I promise it's rhetorical?" No, because "die in a fire" is a bottom-tier, uncreative, stock internet insult. My tiger comment, while admittedly pointed, was at least bespoke. It was also clearly not meant to be serious, I enjoy making jokes. If it wasn't clear then, for the love of God I hope it's clear now. There's a difference between sharp, theatrical hyperbole meant to illustrate a point with some flavor, and just being generically hostile.
On the topic of my grandmother: you seem to think you've found some grand "gotcha." You haven't. You've simply discovered the concept of "scaffolding" in teaching.
Of course I wouldn't dump the entire "fallible but brilliant intern" model on a complete novice in one go. That's not how you explain anything complex. You break it down. The "genius with the world's worst memory" is a facet of the intern model.
I would tell the hypothetical grandma, due to a paucity of my own:
This is an AI. It can talk just like a human, in text or speech. It can even do video one way. (A grandma isn't using Ani from Grok, anime avatars aren't a concern)
It is not actually a human. But you can mostly treat it as a human, if you keep in mind the following:
You need to introduce yourself to it, it knows nothing about you. Think of it as an intern you just met.
It doesn't remember previous conversations by default. There are exceptions, granny, but there not too relevant here.
It's really smart! But it's also forgetful, can make errors, so please double check what it says if it's not more important than ideal dress to Bingo Night.
It can and will flatter you, my what lovely eyes you have granny. Please be careful about whether it's agreeing with you because you're right or because it wants to please you.
It is good at: X, Y and Z. And bad at:.. If you really need A or B, then maybe consider this funny little fella named Claude.
More depending on the context.
That is a perfectly good framework. Now, compare that list of actionable, non-technical advice to the guidance offered by the "stochastic parrot" model. What would that list look like?
This is a parrot.
It just repeats things it's heard without understanding them.
...That's it. That's the whole model.
And this brings me back to your final, bizarre point about me paying for ChatGPT Plus. You think that's a "confession" of bias. I think it's the very foundation of my argument. You don't develop a nuanced, multi-part user model like the one I just laid out by casually playing with the free version. You develop it through deep, sustained use. I was running into the tool's limitations day after day and systematically figuring out the strategies that work. I was doing this well before it was cool.
Plus being a "confession" of bias is a wild misapplication of pop psychology. By that logic, no one who pays for a product can ever be a credible critic or analyst of it. The person who buys a Ford F-150 can't tell you about its turning radius? The person who subscribes to The Economist is just engaging in post-purchase rationalization when they recommend an article? It's absurd.
I pay for it because I use it extensively, both for work and for leisure. That heavy use is precisely why I have a well-developed opinion on its capabilities, its flaws, and the best mental models for using it effectively. It's a credential for my argument, not a disqualifier.
You say you're joking, and then you continue by explaining why you wouldn't intervene in another scenario where you imagine me in a cage with a tiger. You couch your "apparent hostility towards bird fanciers" in the dismissive phrase "quite a bit", leaving yourself wiggle room to continue thinking less of some - like me. Then you tell me, a stranger you have never met and never will meet who lives on the other side of the world, that you don't actually wish me dead. Implying that my concern is for my life, not the insults. Yeah, I know all the tricks, chum.
Do you want to know how I know? Because I used to prioritize my jokes over the rules of the motte. I learned the hard way, through multiple bans, that being clever is no excuse for hostility. And that hostility is often in the eye of the beholder no matter how you meant it to come across.
So where is this line? It's north of blatantly obvious cliched examples of comedic shaming like "die in a fire" that's clear, but apparently south of "I hope you get mauled by a tiger" and "you're dumber than a parrot". How about, "I hope swarms of aphids crawl down your throat"? Or "I almost want to stick a iron hook up your nose and scrape out your brains, but I see there's no point" or maybe "scientists discovered a new sub-atomic particle on the edge of the gluon field - your worthless dick". I really need to know so I can go back to 'joking' people into silence. Either way I'll be damned if I'm going to let a mod get away with it if I can't.
Now, onto your 'scaffolding'. What was it I said you'd have to tell your grandma about your intern?
Huh, looks like I discovered the concept a while ago. And what 'scaffolding' did you just invent? A list of rules that describes an amnesiac, unreliable, potentially flattering (read lying) intern who is bad at certain tasks.
You are still deliberately missing the fundamental concept. Let me try one last time. Cognitive. Shortcut. The goal is to give a novice a powerful, easy to remember tool to 'shortcut' if you will, their biggest barrier - anthropomorphism. Your scaffolding is just a more complicated version of my model. In fact you had to gut your own metaphor (the fallible intern, closer to a human than a parrot) and adopt the primary principle of mine (it's not human) to make it work. It's funny how the grandmas and grandpas I've taught my 'bad' model to have managed to wrap their heads around it immediately - and have gone on to exceed the AI skills of many of my techbro friends.
And as for armchair psychology, you brought up your financial relationship with OpenAI as proof you aren't biased, that you aren't defending the public image of LLMs. I just pointed out how flawed that argument is by explaining basic psychological principles like the sunk cost fallacy. I honestly can not believe a trained psychiatrist is claiming paying for something is proof they aren't biased towards it. It's beyond ridiculous.
And of course paying customers can be credible reviewers. I used to be one for a living. The site I worked for refused to play the '7 out of 10 is the floor' game, so despite being part of the biggest telecommunications network in the country we had to pay for Sega and Xbox Studios games to review them. But we made an effort to check our biases, with each other and our readers. And more importantly, this isn't a product review, this is a slap fight about which mental model is is best for novice AI users. You are heavily invested in your workarounds, I understand. I am heavily invested in mine. And while I haven't been heavily into it since before it was 'cool', I did:
Jump in with both feet. I use Gemini 2.5 pro, which I pay for, every day. I find its g-suite integration to be an incredible efficiency enhancer.
Expand beyond using a single model - I have API credit for DeepSeek, Gemini, Claude, Kimi, ChatGPT, and Grok. I could say I use them every day too, except I'm currently away from my computer.
Develop your nuanced, multi-part user model before you did, with greater clarity.
My amusement at your condescension aside, that makes me biased too. But it also gives me the perspective to know that 'thinking like a GPT power user' isn't a universal solution. And it's working with others that gives me the perspective to know that a simple, portable mental model like the parrot is far more useful for novices across all platforms than a complex personality profile for just one.
I suspect none of what I just said matters though. Much like nothing I've said matters. You aren't arguing to enlighten, you are arguing to win the argument. That's not my assessment, in case you think this is more of my pop psychology, it was the assessment Gemini gave me prior to the last post when I put our conversation into it and asked it how I could possibly get my point across when you hadn't seemed to understand anything I'd said already. I should have listened.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link