Bag of words, have mercy on us

Bag of words, have mercy on us(experimental-history.com)

328 points by ntnbr 161 days ago | 353 comments

bloaf 161 days ago |

Everyone is out here acting like "predicting the next thing" is somehow fundamentally irrelevant to "human thinking" and it is simply not the case.

What does it mean to say that we humans act with intent? It means that we have some expectation or prediction about how our actions will effect the next thing, and choose our actions based on how much we like that effect. The ability to predict is fundamental to our ability to act intentionally.

So in my mind: even if you grant all the AI-naysayer's complaints about how LLMs aren't "actually" thinking, you can still believe that they will end up being a component in a system which actually "does" think.

RayVR 161 days ago | |

Are you a stream of words or are your words the “simplistic” projection of your abstract thoughts? I don’t at all discount the importance of language in so many things, but the question that matters is whether statistical models of language can ever “learn” abstract thought, or become part of a system which uses them as a tool.

My personal assessment is that LLMs can do neither.

ACCount37 161 days ago | | |

Words are the "simplistic" projection of an LLM's abstract thoughts.

An LLM has: words in its input plane, words in its output plane, and A LOT of cross-linked internals between the two.

Those internals aren't "words" at all - and it's where most of the "action" happens. It's how LLMs can do things like translate from language to language, or recall knowledge they only encountered in English in the training data while speaking German.

balamatom 161 days ago | | |

I'm definitely a stream of words.

My "abstract thoughts" are a stream of words too, they just don't get sounded out.

Tbf I'd rather they weren't there in the first place.

But bodies which refuse to harbor an "interiority" are fast-tracked to destruction because they can't suf^W^W^W be productive.

Funny movie scene from somewhere. The sergeant is drilling the troops: "You, private! What do you live for!", and expects an answer along the lines of dying for one's nation or some shit. Instead, the soldier replies: "Well, to see what happens next!"

Davidzheng 161 days ago | | |

Even if they are "simplistic projections", which I don't think is the correct way to think about it, there's no reason that more LLM thoughts in middle layers can't also exist and project down at the end. Though there might be efficency issues because the latent thoughts have to be recomputed a lot.

Though I do think in human brains it's also an interplay where what we write/say also loops back into the thinking as well. Which is something which is efficient for LLMs.

gardenhedge 160 days ago | | |

I am a stream of words - I have even ran out of tokens while speaking before :)

But raising kids, I can clearly see that intelligence isn't just solved by LLMs

akoboldfrying 161 days ago | | |

LLMs and human brains are both just mechanisms. Why would one mechanism a priori be capable of "learning abstract thought", but no others?

If it turns out that LLMs don't model human brains well enough to qualify as "learning abstract thought" the way humans do, some future technology will do so. Human brains aren't magic, special or different.

MyOutfitIsVague 161 days ago | |

> Everyone is out here acting like "predicting the next thing" is somehow fundamentally irrelevant to "human thinking" and it is simply not the case.

Nobody is. What people are doing is claiming that "predicting the next thing" does not define the entirety of human thinking, and something that is ONLY predicting the next thing is not, fundamentally, thinking.

visarga 161 days ago | | |

Well, yes because thinking soon requires interacting, not just ideating. It's in the dialogue between ideation and interaction that we make our discoveries.

agumonkey 160 days ago | | |

when LLM popped out and people started to say 'this is just markov chain on steroid and not thinking' i was a bit confused because a lot of my "thinking" is statistical too.. I very often try to solve an issue by switching a known solution with a different "probable" variant of it (tweaking a parameter)

LLMs have higher dimensions (they map token to grammatical and semantical space) .. it might not be thinking but it seems on its way we're just thinking with more abstractions before producing speech ?... dunno

akoboldfrying 161 days ago | | |

I claim that all of thinking can be reduced to predicting the next thing. Predicting the next thing = thinking in the same way that reading and writing strings of bytes is a universal interface, or every computation can be done by a Turing machine.

Libidinalecon 160 days ago | |

A motorcycle is not "sprinting" and an LLM is not "thinking". Everyone would agree that a motorcycle is not running but the same dumb shit is posted over and over and over on here that somehow the LLM is "thinking".

crazygringo 160 days ago | | |

But your assertion is merely semantic. It doesn't say anything substantive.

I could also say a motorcycle "moves forward" just like a person "moves forward". Whether we use the same or different words for same or different concepts doesn't say anything about the actual underlying similarity.

And please don't call stuff "dumb shit" here. Not appropriate for HN.

Extasia785 160 days ago | | |

A forklift is "lifting" things, despite using a completely different mechanical process as a human "lifting" things. The only real similarity between these kinds of "lifting" is the end result, something is higher up than it was before.

DoctorOetker 160 days ago | | |

is this seriously about continuous rotation versus a pair of double pendulums making a stepping motion?

MattRix 160 days ago | | |

That’s because the motorcycle thing is too simlistic of a comparison. It doesn’t come nearly close to capturing the nuance of the whole LLM “thinking” situation.

efitz 160 days ago | |

AI has made me question what it is to be a human.

I am not having some existential crisis, but if we get to a point where X% of humans cannot outperform “AI” on any task that humans deem “useful”, for some nontrivial value of X, then many assumptions that culture has inculcated into me about humanity are no longer valid.

What is the role of humans then?

Can it be said that humans “think” if they can’t think a thought that a non thinking AI cannot also think?

tjr 160 days ago | | |

If all humans were suddenly wiped off the face of the earth, AI would go silent, and the hardware it runs on would eventually shut down.

If all AI was suddenly wiped off the face of the earth, humans would rebuild it, and would carry on fine in the meantime.

One AI researcher decades ago said something to the effect of: researchers in biology look at living organisms and wonder how they live; researchers in physics look at the cosmos and wonder what all is out there; researchers in artificial intelligence look at computer systems and wonder how they can be made to wonder such things.

throw4847285 160 days ago | | |

But the "AI" is simply a gestalt of all human language. You're looking in a mirror.

zkmon 161 days ago | |

It may be doing the "thinking" and could reach AGI. But we don't want it. We don't want to take a fork lift to the gym. We don't want plastic aliens showing off their AGI and asking humanity to outsource human thinking and decision-making to them.

perrygeo 160 days ago | |

Predicting the next token is not at all the same thing as predicting the next action in a causal chain of actions. Not even close. One is model of language tokens, the other is a model of the physical world. You can come up with all sorts of predictions that can't be expressed cleanly in natural language. And plenty of things that parse cleanly from a language perspective but are unhinged in their description of empirical reality.

voidhorse 161 days ago | |

When you have a thought, are you "predicting the next thing"—can you confidently classify all mental activity that you experience as "predicting the next thing"?

Language and society constrains the way we use words, but when you speak, are you "predicting"? Science allows human beings to predict various outcomes with varying degrees of success, but much of our experience of the world does not entail predicting things.

How confident are you that the abstractions "search" and "thinking" as applied to the neurological biological machine called the human brain, nervous system, and sensorium and the machine called an LLM are really equatable? On what do you base your confidence in their equivalence?

Does an equivalence of observable behavior imply an ontological equivalence? How does Heisenberg's famous principle complicate this when we consider the role observer's play in founding their own observations? How much of your confidence is based on biased notions rather than direct evidence?

The critics are right to raise these arguments. Companies with a tremendous amount of power are claiming these tools do more than they are actually capable of and they actively mislead consumers in this manner.

ctoth 161 days ago | | |

> When you have a thought, are you "predicting the next thing"

Yes. This is the core claim of the Free Energy Principle[0], from the most-cited neuroscientist alive. Predictive processing isn't AI hype - it's the dominant theoretical framework in computational neuroscience for ~15 years now.

> much of our experience of the world does not entail predicting things

Introspection isn't evidence about computational architecture. You don't experience your V1 doing edge detection either.

> How confident are you that the abstractions "search" and "thinking"... are really equatable?

This isn't about confidence, it's about whether you're engaging with the actual literature. Active inference[1] argues cognition IS prediction and action in service of minimizing surprise. Disagree if you want, but you're disagreeing with Friston, not OpenAI marketing.

> How does Heisenberg's famous principle complicate this

It doesn't. Quantum uncertainty at subatomic scales has no demonstrated relevance to cognitive architecture. This is vibes.

> Companies... are claiming these tools do more than they are actually capable of

Possibly true! But "is cognition fundamentally predictive" is a question about brains, not LLMs. You've accidentally dismissed mainstream neuroscience while trying to critique AI hype.

[0] https://www.nature.com/articles/nrn2787

[1] https://mitpress.mit.edu/9780262045353/active-inference/

Ukv 161 days ago | | |

> can you confidently classify all mental activity that you experience as "predicting the next thing"? [...] On what do you base your confidence in their equivalence?

To my understanding, bloaf's claim was only that the ability to predict seems a requirement of acting intentionally and thus that LLMs may "end up being a component in a system which actually does think" - not necessarily that all thought is prediction or that an LLM would be the entire system.

I'd personally go further and claim that correctly generating the next token is already a sufficiently general task to embed pretty much any intellectual capability. To complete `2360 + 8352 * 4 = ` for unseen problems is to be capable of arithmetic, for instance.

Yhippa 161 days ago | | |

Boo LLM-generated comments!

bloaf 161 days ago | | |

> When you have a thought, are you "predicting the next thing"—can you confidently classify all mental activity that you experience as "predicting the next thing"?

So notice that my original claim was "prediction is fundamental to our ability to act with intent" and now your demand is to prove that "prediction is fundamental to all mental activity."

That's a subtle but dishonest rhetorical shift to make me have to defend a much broader claim, which I have no desire to do.

> Language and society constrains the way we use words, but when you speak, are you "predicting"?

Yes, and necessarily so. One of the main objections that dualists use to argue that our mental processes must be immaterial is this [0]:

* If our mental processes are physical, then there cannot be an ultimate metaphysical truth-of-the-matter about the meaning of those processes.

* If there is no ultimate metaphysical truth-of-the-matter about what those processes mean, then everything they do and produce are similarly devoid of meaning.

* Asserting a non-dualist mind therefore implies your words are meaningless, a self-defeating assertion.

The simple answer to this dualist argument is precisely captured by this concept of prediction. There is no need to assert some kind of underlying magical meaning to be able to communicate. Instead, we need only say that in the relevant circumstances, our minds are capable of predicting what impact words will have on the receiver and choosing them accordingly. Since we humans don't have access to each other's minds, we must not learn these impacts from some kind of psychic mind-to-mind sense, but simply from observing the impacts of the words we choose on other parties; something that LLMs are currently (at least somewhat) capable of observing.

[0] https://www.newdualism.org/papers/E.Feser/Feser-acpq_2013.pd...

If you read the above link you will see that they spell out 3 problems with our understanding of thought:

Consciousness, intentionality, and rationality.

Of these, I believe prediction is only necessary for intentionality, but it does have some roles to play in consciousness and rationality.

micromacrofoot 160 days ago | |

Yes, personally I'm completely fine with the fact that LLMs don't actually think. I don't care that they're not AGI, though the hysterics about "AGI is so close now" seems silly to me. Fusion reactors and self-driving cars are just around the same corner.

They prove to have some useful utility to me regardless.

gamerDude 161 days ago | |

I'm an LLMs are being used in workflows they don't make sense in-sayer. And while yes, I can believe that LLMs can be part of a system that actually does think, I believe that to achieve true "thinking", it would likely be a system that is more deterministic in its approach rather than probabilistic.

Especially when modeling acting with intent. The ability to measure against past results and think of new innovative approaches seems like it may come from a system that may model first and then use LLM output. Basically something that has a foundation of tools rather than an LLM using MCP. Perhaps using LLMs to generate a response that humans like to read, but not in them coming up with the answer.

Either way, yes, its possible for a thinking system to use LLMs (and potentially humans piece together sentences in a similar way), but its also possible LLMs will be cast aside and a new approach will be used to create an AGI.

So for me: even if you are an AI-yeasayer, you can still believe that they won't be a component in an AGI.

visarga 161 days ago | | |

You can make a separate model for the task, which is based on well chosen features and calibrated from actual data. Then the LLM only needs to generate the arguments to this model (extract those features from messages) and call it like a MCP tool. This external tool can be a simple Sklearn model.

jampekka 161 days ago | |

A good heuristic is that if an argument resorts to "actually not doing <something complex sounding>" or "just doing <something simple sounding>" etc, it is not a rigorous argument.

bamboozled 161 days ago | |

The issue is that prediction is "part" of the human thought process, it's not the full story...

bloaf 161 days ago | | |

And the big players have built a bunch of workflows which embed many other elements besides just "predictions" into their AI product. Things like web search, to incorporating feedback from code testing, to feeding outputs back into future iterations. Who is to say that one or more of these additions has pushed the ensemble across the threshold and into "real actual thinking."

The near-religious fervor which people insist that "its just prediction" makes me want to respond with some religious allusions of my own:

> Who is this that wrappeth up sentences in unskillful words? Gird up thy loins like a man: I will ask thee, and answer thou me. Where wast thou when I laid up the foundations of the earth? tell me if thou hast understanding. Who hath laid the measures thereof, if thou knowest? or who hath stretched the line upon it?

The point is that (as far as I know) we simply don't know the necessary or sufficient conditions for "thinking" in the first place, let alone "human thinking." Eventually we will most likely arrive at a scientific consensus, but as of right now we don't have the terms nailed down well enough to claim the kind of certainty I see from AI-detractors.

throwaway150 161 days ago | | |

> The issue is that prediction is "part" of the human thought process, it's not the full story...

Do you have a proof for this?

Surely such a profound claim about human thought process must have a solid proof somewhere? Otherwise who's to say all of human thought process is not just a derivative of "predicting the next thing"?

observationist 160 days ago | |

It's fascinating when you look at each technical component of cognition in human brains and contrast against LLMs. In humans, we have all sorts of parallel asynchronous processes running, with prediction of columnar activations seemingly the fundamental local function, with tens of thousands of mini columns and regions in the brain corresponding to millions of networked neurons using the "predict which column fires next" objective to increment or decrement the relative contribution of any functional unit.

In the case of LLMs you run into similarities, but they're much more monolithic networks, so the aggregate activations are going to scan across billions of neurons each pass. The sub-networks you can select each pass by looking at a threshold of activations resemble the diverse set of semantic clusters in bio brains - there's a convergent mechanism in how LLMs structure their model of the world and how brains model the world.

This shouldn't be surprising - transformer networks are designed to learn the complex representations of the underlying causes that bring about things like human generated text, audio, and video.

If you modeled a star with a large transformer model, you would end up with semantic structures and representations that correlate to complex dynamic systems within the star. If you model slug cellular growth, you'll get structure and semantics corresponding to slug DNA. Transformers aren't the end-all solution - the paradigm is missing a level of abstraction that fully generalizes across all domains, but it's a really good way to elicit complex functions from sophisticated systems, and by contrasting the way in which those models fail against the way natural systems operate, we'll find better, more general methods and architectures, until we cross the threshold of fully general algorithms.

Biological brains are a computational substrate - we exist as brains in bone vats, connected to a wonderfully complex and sophisticated sensor suite and mobility platform that feeds electrically activated sensory streams into our brains, which get processed into a synthetic construct we experience as reality.

Part of the underlying basic functioning of our brains is each individual column performing the task of predicting which of any of the columns it's connected to will fire next. The better a column is at predicting, the better the brain gets at understanding the world, and biological brains are recursively granular across arbitrary degrees of abstraction.

LLMs aren't inherently incapable of fully emulating human cognition, but the differences they exhibit are expensive. It's going to be far more efficient to modify the architecture, and this may diverge enough that whatever the solution ends up being, it won't reasonably be called an LLM. Or it might not, and there's some clever tweak to things that will push LLMs over the threshold.

moralIsYouLie 161 days ago | |

most humans in any percentile act towards the thing of someone else. most of these things are a lot worse than what the human "would originally intend". this behavior stems from 100s and thousands of nudges since childhood.

the issue with AI and AI-naysayers is, by analogy, this: cars were build to drive from A to Z. people picked up tastes and some people started building really cool looking cars. the same happens on the engineering side. then portfolio communists came with their fake capitalism and now cars are build to drive over people but don't really work because people, thankfully, are overwhelming still fighting to attempt to act towards their own intents.

Nevermark 161 days ago | |

Exactly. Our base learning is by example, which is very much learning to predict.

Predict the right words, predict the answer, predict when the ball bounces, etc. Then reversing predictions that we have learned. I.e. choosing the action with the highest prediction of the outcome we want. Whether that is one step, or a series of predicted best steps.

Also, people confuse different levels of algorithm.

There are at least 4 levels of algorithm:

• 1 - The architecture.

This input-output calculation for pre-trained models are very well understood. We put together a model consisting of matrix/tensor operations and few other simple functions, and that is the model. Just a normal but high parameter calculation.

• 2 - The training algorithm.

These are completely understood.

There are certainly lots of questions about what is most efficient, alternatives, etc. But training algorithms harnessing gradients and similar feedback are very clearly defined.

• 3 - The type of problem a model is trained on.

Many basic problem forms are well understood. For instance, for prediction we have an ordered series of information, with later information to be predicted from earlier information. It could simply be an input and response that is learned. Or a long series of information.

• 4 - The solution learned to solve (3) the outer problem, using (2) the training algorithm on (1) the model architecture.

People keep confusing (4) with (1), (2) or (3). But it is very different.

For starters, in the general case, and for most any challenging problem, we never understand their solution. Someday it might be routine, but today we don't even know how to approach that for any significant problem.

Secondly, even with (1), (2), and (3) exactly the same, (4) is going to be wildly different based on the data characterizing the specific problem to solve. For complex problems, like language, layers and layers of sub-solutions to sub-problems have to be solved, and since models are not infinite in size, ways to repurpose sub-solutions, and weave together sub-solutions to address all the ways different sub-problems do and don't share commonalities.

Yes, prediction is the outer form of their solution. But to do that they have to learn all the relationships in the data. And there is no limit to how complex relationships in data can be. So there is no limit on the depths or complexity of the solutions found by successfully trained models.

Any argument they don't reason, based on the fact that they are being trained to predict, confuses at least (3) and (4). That is a category error.

It is true, they reason a lot more like our "fast thinking", intuitive responses, than our careful deep and reflective reasoning. And they are missing important functions, like a sense of what they know or don't. They don't continuously learn while inferencing. Or experience meta-learning, where they improve on their own reasoning abilities with reflection, like we do. And notoriously, by design, they don't "see" the letters that spell words in any normal sense. They see tokens.

Those reasoning limitations can be irritating or humorous. Like when a model seems to clearly recognize a failure you point out, but then replicates the same error over and over. No ability to learn on the spot. But they do reason.

Today, despite many successful models, nobody understands how models are able to reason like they do. There is shallow analysis. The weights are there to experiment with. But nobody can walk away from the model and training process, and build a language model directly themselves. We have no idea how to independently replicate what they have learned, despite having their solution right in front of us. Other than going through the whole process of retraining another one.

nottorp 160 days ago | |

This is the "but LLMs will get better, trust me" thread?

sublinear 161 days ago | |

LLMs merely interpolate between the feeble artifacts of thought we call language.

The illusion wears off after about half an hour for even the most casual users. That's better than the old chatbots, but they're still chatbots.

Did anyone ever seriously buy the whole "it's thinking" BS when it was Markov chains? What makes you believe today's LLMs are meaningfully different?

stavros 161 days ago | | |

Did anyone ever seriously buy the whole "it's transporting" BS when it was wheelbarrows? What makes you believe today's trucks are meaningfully different?

mapontosevenths 160 days ago | |

I suspect that people instinctively believe they have free will, both because it feels like we do, and because society requires us to behave that way even when we don't.

The truth is that the evidence says we don't. See the Libet experiment and its many replications.

Your decisions can be predicted from brain scans up to 10 seconds before you make them, which means they are as deterministic as an LLM's. Sorry, I guess.

Hendrikto 160 days ago | | |

> Your decisions can be predicted from brain scans up to 10 seconds before you make them, which means they are as deterministic as an LLM's.

This conclusion does not follow from the result at all.

namero999 160 days ago | | |

Libet has only measured the latency of metaconsciousness/cognition, nothing else. It says nothing about free will, which is ill defined anyway.

beepbooptheory 160 days ago | | |

What is the import of this to you here? Whether you have free will or you feel like you do, kinda same difference for this particular point right? It doesn't make me more human actually having free will, it is sufficient to simply walk around as if I do.

But beyond that, what do you want to say here? What is lost, what is gained? Are you wanting to say this makes us more like an LLM? How so?

jnd-cz 160 days ago | | |

I looked up the Libet experiment:

"Implications

The experiment raised significant questions about free will and determinism. While it suggested that unconscious brain activity precedes conscious decision-making, Libet argued that this does not negate free will, as individuals can still choose to suppress actions initiated by unconscious processes."

viccis 161 days ago |

Every day I see people treat gen AI like a thinking human, Dijkstra's attitudes about anthropomorphizing computers is vindicated even more.

That said, I think the author's use of "bag of words" here is a mistake. Not only does it have a real meaning in a similar area as LLMs, but I don't think the metaphor explains anything. Gen AI tricks laypeople into treating its token inferences as "thinking" because it is trained to replicate the semiotic appearance of doing so. A "bag of words" doesn't sufficiently explain this behavior.

palata 161 days ago |

Slightly unfortunate that "Bag of words" is already a different concept: https://en.wikipedia.org/wiki/Bag_of_words.

My second thought is that it's not the metaphor that is misleading. People have been told thousands of times that LLMs don't "think", don't "know", don't "feel", but are "just a very impressive autocomplete". If they still really want to completely ignore that, why would they suddenly change their mind with a new metaphor?

Humans are lazy. If it looks true enough and it cost less effort, humans will love it. "Are you sure the LLM did your job correctly?" is completely irrelevant: people couldn't care less if it's correct or not. As long as the employer believes that the employee is "doing their job", that's good enough. So the question is really: "do you think you'll get fired if you use this?". If the answer is "no, actually I may even look more productive to my employer", then why would people not use it?

kaycebasques 160 days ago | |

> Slightly unfortunate that "Bag of words" is already a different concept

Yes, subconsciously I kept trying to map this article's ideas to word2vec and continuous-bag-of-words.

4bpp 161 days ago |

As usual with these, it helps to try to keep the metaphor used for downplaying AI, but flip the script. Let's grant the author's perception that AI is a "bag of words", which is already damn good at producing the "right words" for any given situation, and only keeps getting better at it.

Sure, this is not the same as being a human. Does that really mean, as the author seems to believe without argument, that humans need not be afraid that it will usurp their role? In how many contexts is the utility of having a human, if you squint, not just that a human has so far been the best way to "produce the right words in any given situation", that is, to use the meat-bag only in its capacity as a word-bag? In how many more contexts would a really good magic bag of words be better than a human, if it existed, even if the current human is used somewhat differently? The author seems to rest assured that a human (long-distance?) lover will not be replaced by a "bag of words"; why, especially once the bag of words is also ducttaped to a bag of pictures and a bag of sounds?

I can just imagine someone - a horse breeder, or an anthropomorphised horse - dismissing all concerns on the eve of the automotive revolution, talking about how marketers and gullible marks are prone to hippomorphising anything that looks like it can be ridden and some more, and sprinkling some anecdotes about kids riding broomsticks, legends of pegasi and patterns of stars in the sky being interpreted as horses since ancient times.

tristanlukens 161 days ago |

> If we allow ourselves to be seduced by the superficial similarity, we’ll end up like the moths who evolved to navigate by the light of the moon, only to find themselves drawn to—and ultimately electrocuted by—the mysterious glow of a bug zapper.

Woah, that hit hard

bitwize 161 days ago |

I was trying to explain the concept of "token prediction" to my wife, whose eyes glaze over when discussing such technical topics. (I think she has the brainpower to understand them, but a horrible math teacher gave her a taste aversion to even attempting to that hasn't gone away. So she just buys Apple stuff and hopes Tim Apple hasn't shuffled around the UI bits AGAIN.)

I stumbled across a good-enough analogy based on something she loves: refrigerator magnet poetry, which if it's good consists of not just words but also word fragments like "s", "ed", and "ing" kinda like LLM tokens. I said that ChatGPT is like refrigerator magnet poetry in a magical bag of holding that somehow always gives the tile that's the most or nearly the most statistically plausible next token given the previous text. E.g., if the magnets already up read "easy come and easy ____", the bag would be likely to produce "go". That got into her head the idea that these things operate based on plausibility ratings from a statistical soup of words, not anything in the real world nor any internal cogitation about facts. Any knowledge or thought apparent in the LLM was conducted by the original human authors of the words in the soup.

CamperBob2 161 days ago | |

Did you explain how LLMs can achieve gold-medal performance at math competitions involving original problems, without any original knowledge or thought?

Did she ask if a "statistical soup of words," if large enough, might somehow encode or represent something a little more profound than just a bunch of words?

AlexeyBelov 158 days ago | | |

Objection. Leading questions.

tkgally 161 days ago |

I am unsure myself whether we should regard LLMs as mere token-predicting automatons or as some new kind of incipient intelligence. Despite their origins as statistical parrots, the interpretability research from Anthropic [1] suggests that structures corresponding to meaning do exist inside those bundles of numbers and that there are signs of activity within those bundles of numbers that seem analogous to thought.

That said, I was struck by a recent interview with Anthropic’s Amanda Askell [2]. When she talks, she anthropomorphizes LLMs constantly. A few examples:

“I don't have all the answers of how should models feel about past model deprecation, about their own identity, but I do want to try and help models figure that out and then to at least know that we care about it and are thinking about it.”

“If you go into the depths of the model and you find some deep-seated insecurity, then that's really valuable.”

“... that could lead to models almost feeling afraid that they're gonna do the wrong thing or are very self-critical or feeling like humans are going to behave negatively towards them.”

[1] https://www.anthropic.com/research/team/interpretability

[2] https://youtu.be/I9aGC6Ui3eE

djoldman 161 days ago |

As a consequence of my profession, I understand how LLMs work under the hood.

I also know that we data and tech folks will probably never win the battle over anthropomorphization.

The average user of AI, nevermind folks who should know better, is so easily convinced that AI "knows," "thinks," "lies," "wants," "understands," etc. Add to this that all AI hosts push this perspective (and why not, it's the easiest white lie to get the user to act so that they get a lot of value), and there's really too much to fight against.

We're just gonna keep on running into this and it'll just be like when you take chemistry and physics and the teachers say, "it's not actually like this but we'll get to how some years down the line- just pretend this is true for the time being."

IAmBroom 160 days ago |

In this thread: 99% of posters using their own personal definition of "thinking" without explaining it; 0.99% of posters complaining that it all depends on what that definition is; not enough posts yet for that 0.01% response to occur...

yannyu 160 days ago | |

There's no definition of thinking that isn't a purely internal phenomenon, which means that there's no way to point a diagnostic device at someone and determine whether they're thinking. The only way to determine whether something is conscious/thinking is through some sort of inference, which is why Turing landed on the Turing Test that he did. Problem is, technology over the past 5 years pretty easily passes variations of the Turing Test, and exposed a lot of its limits as well.

So the next definition of detecting "thinking" will have to be externally observable and inferrable like a Turing Test, but get into the other things that we consider part of consciousness/thinking.

Often this is some combination of introspection (understanding internal states), perception (understanding external objects), and synthesis of the two into testable hypotheses in some sort of feedback loop between the internal representation of the world and the external feedback from the world.

Right now, a chatbot can say all sorts of things about itself and about the world, but none of that is based on real-time, factual information. Whereas an animal can't speak, but they clearly process information and consider it when determining their future and current actions.

rdiddly 160 days ago | |

It's not obvious to me what you expect from this hypothetical 0.01% post, or in other words, what about it makes it a one-in-ten-thousand post?

raincole 161 days ago |

> “Bag of words” is a also a useful heuristic for predicting where an AI will do well and where it will fail. “Give me a list of the ten worst transportation disasters in North America” is an easy task for a bag of words, because disasters are well-documented. On the other hand, “Who reassigned the species Brachiosaurus brancai to its own genus, and when?” is a hard task for a bag of words, because the bag just doesn’t contain that many words on the topic

It is... such a retrospective narrative. It's so obvious that the author learned about this example first than came with the reasoning later, just to fit in his view of LLM.

Imaging if ChatGPT answered this question correctly. Would that change the author's view? Of course not! They'll just say:

> “Bag of words” is a also a useful heuristic for predicting where an AI will do well and where it will fail. Who reassigned the species Brachiosaurus brancai to its own genus, and when?” is an easy task for a bag of words, because the information has appeared in the words it memorizes.

I highly doubt this author has predicted that "bag of Words" can do image editing before OpenAI released that.

Kim_Bruning 161 days ago |

This is essentially Lady Lovelace's objection from the 19th century [1]. Turing addressed this directly in "Computing Machinery and Intelligence" (1950) [2], and implicitly via the halting problem in "On Computable Numbers" (1936) [3]. Later work on cellular automata, famously Conway's Game of Life [4], demonstrates more conclusively that this framing fails as a predictive model: simple rules produce structures no one "put in."

A test I did myself was to ask Claude (The LLM from Anthropic) to write working code for entirely novel instruction set architectures (e.g., custom ISAs from the game Turing Complete [5]), which is difficult to reconcile with pure retrieval.

[1] Lovelace, A. (1843). Notes by the Translator, in Scientific Memoirs Vol. 3. ("The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform.") Primary source: https://en.wikisource.org/wiki/Scientific_Memoirs/3/Sketch_o.... See also: https://www.historyofdatascience.com/ada-lovelace/ and https://writings.stephenwolfram.com/2015/12/untangling-the-t...

[2] https://academic.oup.com/mind/article/LIX/236/433/986238

[3] https://www.cs.virginia.edu/~robins/Turing_Paper_1936.pdf

[4] https://web.stanford.edu/class/sts145/Library/life.pdf

[5] https://store.steampowered.com/app/1444480/Turing_Complete/

ares623 161 days ago |

I think a better metaphor is the Library of Babel.

A practically infinite library where both gibberish and truth exist side by side.

The trick is navigating the library correctly. Except in this case you can’t reliably navigate it. And if you happen to stumble upon some “future truth” (i.e. new knowledge), you still need to differentiate it from the gibberish.

So a “crappy” version of the Library of Babel. Very impressive, but the caveats significantly detract from it.

dearing 160 days ago | |

This is where I sit too. Obviously language is an expression of thought but the Library of Babel is a great example that language without intent is just garbage. You got me thinking of reading before the internet. You'd grab a book and internalize the subject, later refining over time with more books, experiments and other forms of conversation. That journey of developing your own model is undervalued in understanding. That first book could of be absolute shit but you couldn't know that.

I've been learning more about roses lately and the amount of information on them varies so much because the world roses live in is equally varied. LLMs make for a better search engine but you still need to develop your own internal models, worse yet - if LLMs continue to be refined off of cul-de-sac conclusions then all the wisdom of the journey is lost both to the consumer and the LLM itself.

globular-toast 161 days ago | |

It's like a highly compressed version of the Library. You're basically trying to discern real details from compression artifacts.

ares623 161 days ago | | |

And the halls and shelves keep shuffling around randomly.

d4rkn0d3z 161 days ago |

An LLM creates a high fidelity statistical probabistic model of human language. The hope is to capture the input/output of various hierarchical formal and semiformal systems of logic that transit from human to human, which we know as "Intelligence".

Unfortunately, its corpus is bound to contain noise/nonsense that follows no formal reasoning system but contributes to the ill advised idea that an AI should sound like a human to be considered intelligent. Therefore it is not a bag of words but a bag of probabilities perhaps. This is important because the fundamental problem is that an LLM is not able, by design, to correctly model the most fundamental precept of human reason, namely the law of non-contradiction. An LLM must, I repeat must assign nonvanishing probability to both sides of a contradiction, and what's worse is the winning side loses, since long chains of reason are modelled with probability the longer the chain, the less likely an LLM is to follow it. Moreover, whenever there is actual debate on an issue such that the corpus is ambiguous the LLM becomes chaotic, necessarily, on that issue.

I literally just had an AI prove the forgoing with some rigor, and in the very next prompt, I asked it to check my logical reasoning for consistency and it claimed it was able to do so (->|<-).

A4ET8a8uTh0_v2 161 days ago | |

^^; I think this post is close to singularity as we may get on this Monday.

tibbar 161 days ago |

The problem with these metaphors is that they don't really explain anything. LLMs can solve countless problems today that we would have previously said were impossible because there are not enough examples in the training data. (EG, novel IMO/ICPC problems.) One way that we move the goal posts is to increase the level of abstraction: IMO/ICPC problems are just math problems, right? There are tons of those in the data set!

But the truth is there has been a major semantic shift. Previously LLMs could only solve puzzles whose answers were literally in the training data. It could answer a math puzzle it had seen before, but if you rephrased it only slightly it could no longer answer.

But now, LLMs can solve puzzles where, like, it has seen a certain strategy before. The newest IMO and ICPC problems were only "in the training data" for a very, very abstract definition of training data.

The goal posts will likely have to shift again, because the next target is training LLMs to independently perform longer chunks of economically useful work, interfacing with all the same tools that white-collar employees do. It's all LLM slop til it isn't, same as the IMO or Putnam exam.

And then we'll have people saying that "white collar employment was all in the training data anyway, if you think about it," at which point the metaphor will have become officially useless.

FarmerPotato 161 days ago | |

I see a lesson in how both metaphors don't explain it. Bag-of-words metaphor is ridiculous, but shows us the absurdity of the first metaphor.

tibbar 161 days ago | | |

Yes, there are really two parallel claims here, aren't there: LLMs are not people (true, maybe true forever), and LLMs are only good at things that are well-represented in text form already. (false in certain categories and probably expanding to more in the future.)

voidhorse 161 days ago |

The defenders and the critics around LLM anthropomorphism are both wrong.

The defenders are right insofar as the (very loose) anthropomorphizing language used around LLMs is justifiable to the extent that human beings also rely on disorder and stochastic processes for creativity. The critics are right insofar as equating these machines to humans is preposterous and mostly relies on significantly diminishing our notion of what "human" means.

Both sides fail to meet the reality that LLMs are their own thing, with their own peculiar behaviors and place in the world. They are not human and they are somewhat more than previous software and the way we engage with it.

However, the defenders are less defensible insofar as their take is mostly used to dissimulate in efforts to make the tech sound more impressive than it actually is. The critics at least have the interests of consumers and their full education in mind—their position is one that properly equips consumers to use these tools with an appropriate amount of caution and scrutiny. The defenders generally want to defend an overreaching use of metaphor to help drive sales.

jrm4 160 days ago |

I'm partial to the metaphor I made up:

They are search engines that can remix results.

I like this one because I think most modern folks have a usefully accurate model of what a search engine is in their heads, and also what "remixing" is, which adds up to a better metaphor than "human machine" or whatever.

FatherOfCurses 160 days ago |

A few years ago they made the Cloud-to-Butt browser plugin to ridicule the overuse of cloud concepts.

I would heartily embrace an "AI-to-Bag of Words" browser plugin.

cowsandmilk 161 days ago |

Title is confusing given https://en.wikipedia.org/wiki/Bag-of-words_model

But even more than that, today’s AI chats are far more sophisticated than probabilistically producing the next word. Mixture of experts routes to different models. Agents are able to search the web, write and execute programs, or use other tools. This means they can actively seek out additional context to produce a better answer. They also have heuristics for deciding if an answer is correct or if they should use tools to try to find a better answer.

The article is correct that they aren’t humans and they have a lot of behaviors that are not like humans, but oversimplifying how they work is not helpful.

jrowen 161 days ago |

The bag of words reminds me of the Chinese room.

"The machine accepts Chinese characters as input, carries out each instruction of the program step by step, and then produces Chinese characters as output. The machine does this so perfectly that no one can tell that they are communicating with a machine and not a hidden Chinese speaker.

The questions at issue are these: does the machine actually understand the conversation, or is it just simulating the ability to understand the conversation? Does the machine have a mind in exactly the same sense that people do, or is it just acting as if it had a mind?"

https://en.wikipedia.org/wiki/Chinese_room

Kim_Bruning 161 days ago | |

Chinese room has been discussed to death of course.

Here's one fun approach (out of 100s) :

What if we answer the Chinese room with the Systems Reply [1]?

Searle countered the systems reply by saying he would internalize the Chinese room.

But at that point it's pretty much exactly the Cartesian theater[2] : with room, homunculus, implement.

But the Cartesian theater is disproven, because we've cut open brains and there's no room in there to fit a popcorn concession.

[1] https://plato.stanford.edu/entries/chinese-room/

[2] https://en.wikipedia.org/wiki/Cartesian_theater

jrowen 161 days ago | | |

It just seemed like relevant background that the author might not have been aware of, adjacent and substantial enough to warrant a mention.

I think there is some validity to the Cartesian theater, in that the whole of the experience that we perceive with our senses is at best an interpretation of a projection or subset of "reality."

morpheos137 160 days ago |

Thinking can not be separated from motivation. It's really simple. Humans and other organisms fundamentally think to replicate their DNA. Until AI has a similar incentive structure driving it, it won't be thinking. There is no human behavior or thought that can not be explained by evolutionary drives. It is really perplexing to me how people think "intelligence" is some kind of concrete thing that just magically emerges from a certain degree of computational complexity. I argue instead that intelligence is an adaptive behavior emerging from evolutionary drives interacting with the real world. World models are not prerequisite but consequent of such molded apparatus. Machines won't become intelligent until it is adaptive for them to do so. There is no magic just evolutionary drives and physical possibility. Our current top down approach of "pre-training" LLMs is bound to fail because it does not allow for real time emergence of adaptive behaviors such as general intelligence. Mimicking intelligence through predicting the next word is no more intelligence than a photograph of something is an actual thing. Training a combinatorial network to interpolate images and words is not the same thing as adaptive self modifying behavior in the real world of physics such as organisms engage with through the set of behaviors that we call intelligence.

coppsilgold 161 days ago |

Is a brain not a token prediction machine?

Tokens in form of neural impulses go in, tokens in the form of neural impulses go out.

We would like to believe that there is something profound happening inside and we call that consciousness. Unfortunately when reading about split-brain patient experiments or agenesis of the corpus callosum cases I feel like we are all deceived, every moment of every day. I came to realization that the confabulation that is observed is just a more pronounced effect of the normal.

MyOutfitIsVague 161 days ago | |

Could an LLM trained on nothing and looped upon itself eventually develop language, more complex concepts, and everything else, based on nothing? If you loop LLMs on each other, training them so they "learn" over time, will they eventually form and develop new concepts, cultures, and languages organically over time? I don't have an answer to that question, but I strongly doubt it.

There's clearly more going on in the human mind than just token prediction.

coppsilgold 161 days ago | | |

If you come up with a genetic algorithm scaffolding to affect both the architecture and the training algorithm, and then you instantiate it in an artificial selection environment, and you also give it trillions generations to evolve evolvability just right (as life had for billions of years) then the answer is yes, I'm certain it will and probably much sooner than we did.

Also, I think there is a very high chance that given an existing LLM architecture there exists a set of weights that would manifest a true intelligence immediately upon instantiation (with anterograde amnesia). Finding this set of weights is the problem.

protocolture 161 days ago | |

> Is a brain not a token prediction machine?

I would say that, token prediction is one of the things a brain does. And in a lot of people, most of what it does. But I dont think its the whole story. Possibly it is the whole story since the development of language.

jimbokun 161 days ago | |

We know that consciousness exists because we constantly experience it. It’s really the only thing we can ever know with certainty.

That’s the point of “I think therefore I am.”

danielbln 161 days ago | | |

You know that your own consciousness exists, that's where certainty ends. The rest of us might just pretend. :)

layer8 161 days ago |

Ugly giant bags of mostly words are easy to confuse with ugly giant bags of mostly water.

emsign 161 days ago |

  But we don’t go to baseball games, spelling bees, and
  Taylor Swift concerts for the speed of the balls, the
  accuracy of the spelling, or the pureness of the
  pitch. We go because we care about humans doing those
  things. It wouldn’t be interesting to watch a bag of
  words do them—unless we mistakenly start treating
  that bag like it’s a person.unless we mistakenly
  start treating that bag like it’s a person.

That seems to be the marketing strategy of some very big, now AI dependend companies. Sam Altman and others exaggerating and distorting the capabilities and future of AI.

The biggest issue when it comes to AI is still the same truth as with other technology. It's important who controls it. Attributing agency and personality to AI is a dangerous red flag.

nephihaha 161 days ago | |

A lot of us wouldn't go to a Taylor Swift concert. I had to endure several days of interrupted commuting thanks to them though.

Support alternative and independent bands. They're around, and many are enjoyable. (Some are not but avoid them LOL.)

torben-friis 161 days ago |

I’ve made this point several times: sure, an anthropomorphized LLM is misleading, but would you rather have them seem academic?

At least the human tone implies fallibility, you don’t want them acting like interactive Wikipedia.

andai 161 days ago | |

It's a concussed savant with anretrograde amnesia in a hyperbolic time chamber.

binary132 161 days ago | |

Yes I would VERY much prefer that they not use that awful casual drivel.

danielbln 161 days ago | | |

So configure your LLM of choice to not.

jimbokun 161 days ago |

Best quote from the article:

> That’s also why I see no point in using AI to, say, write an essay, just like I see no point in bringing a forklift to the gym. Sure, it can lift the weights, but I’m not trying to suspend a barbell above the floor for the hell of it. I lift it because I want to become the kind of person who can lift it. Similarly, I write because I want to become the kind of person who can think.

altmanaltman 161 days ago | |

I don't really like the assumption that anyone who uses AI to, say, write an essay, is not the "kind of person who can think."

And using AI to replace things you find recreational is not the point. If you got paid $100 each time you lifted a weight, would you see a point in bringing a forklift to the gym if it's allowed? Or will that make you a person who is so dumb that they cannot think, as the author is implying?

lotyrin 161 days ago | | |

As capable as they get, I still don't see a lot of uses for these things, myself, still. Sometimes if I'm fundamentally uninspired I'll have a model roll the dice, decide what I do or don't like about where it went to create a sense of momentum, but that's the limit. There's never any of its output in my output, even in spirit unless it managed to go somewhere inspiring, it's just a way to let me warm up my generation and discrimination muscles. "Someone is wrong on the internet"-as-a-service, basically.

Generally, if I come across an opportunity to produce ideas or output, I want to capitalize on it for growing my skills and produce an individual and authentic artistic expression where I want to have very fine control over the output in a way that prompt-tweak-verify simply cannot provide.

I don't value the parts it fills in which weren't intentional on the part of the prompter, just send me your prompt instead. I'd rather have a crude sketch and a description than a high fidelity image that obscures them.

But I'm also the kind of person that never enjoyed manufactured pop music or blockbusters unless there's a high concept or technical novelty in addition to the high budget, generally prefer experimental indie stuff, so maybe there's something I just can't see.

klipt 161 days ago | | |

The same person could use a forklift at work, and lift weights manually at the gym.

Just pick the right tool for the job: don't take the forklift into the gym, and don't try to overhead press thousands of pounds that would fracture your spine.

jimbokun 160 days ago | | |

I notice you make no concrete defense of the value of having an AI write an essay for you.

Aloha 161 days ago | |

If you're writing an essay to prove you can or to speak your words - then you should do it yourself - but sometimes you just need an essay to summarize a complex topic as a deliverable.

monegator 161 days ago | |

tough most people either don't get it or are lay people that do not want to become the kind of people who can think. I go with the second one

b112 161 days ago | | |

Russ Hanneman's thigh implants are a key example. Appearances are all to some people. Actual growth is meaningless to them.

The problem with AI, is that they waste the time of dedicated, thinking humans which care to improve themselves. If I write a three paragraph email on a technical topic, and some yahoo responds with AI, I'm now responding to gibberish.

The other side may not have read, may not understand, and is just interacting to save time. Now my generous nature, which is to help others and interact positively, is being wasted to reply to someone who seems to have put thought and care into a response, but instead was just copying and pasting what something else output.

We have issues with crackers on the net. We have social media. We have political interference. Now we have humans pretending to interact, rendering online interactions even more silly and harmful.

If this trend continues, we'll move back to live interaction just to reduce this time waste.

acituan 161 days ago | | |

If the motivation structure is there I don’t see an inherent reason for people to refuse cultivating themselves. Going with the gym analogy lay people did not need gyms when physical work was the norm, cultivation was readily accomplished.

If anything there is a competing motivational structure in which people are incentivized not to think but to consume, react, emote etc. Information processing skills of the individual being deliberately eroded/hijacked/bypassed is not a AI thing. The most obvious example is ads. Thinkers are simply not good for business.

startupsfail 161 days ago | |

Below is the worst quote... It is plain wrong to see an LLM as a bags of words. LLMs pre-trained on large datasets of text are world models. LLMs post-trained with RL are RL-agents that use these modeling capabilities.

> We are in dire need of a better metaphor. Here’s my suggestion: instead of seeing AI as a sort of silicon homunculus, we should see it as a bag of words.

patrickmay 160 days ago | | |

LLMs aren't world models, they are language models. It will be interesting to see which of the LLM implementation techniques will be useful in building world models, but that's not what we have now.

b112 161 days ago | | |

When you see a dog, or describe the entity, do you discuss the genetic makeup or the bone structure?

No, you describe the bark.

The end result is what counts. Training or not, it's just spewing predictive, relational text.

codeulike 161 days ago |

Here’s my suggestion: instead of seeing AI as a sort of silicon homunculus, we should see it as a bag of words.

The best way to think about LLMs is to think of them as a Model of Language, but very Large

zkmon 161 days ago |

But the issue is, 99.999% of the humans won't see is as a bag of words. Because it is easier to go by instincts and see it as a person and assume that it actually knows about magic tricks, can invent new science or theory of everything, and can solve all world problems. Back in the 90's or early 2000's I have seen people writing poems praying and seeking blessings from the Google goddess. People are insanely greedy and instinct-driven. Given this truth, what's the fall-out?

hermitcrab 160 days ago |

"People who experience sleep paralysis sometimes hallucinate a demon-like creature sitting on their chest"

Interestingly, the experience of sleep paralysis seems to change with the culture. Previously, people experienced it as being ridden by a night hag or some other malevolent supernatural being. More recently, it might account for many supposed alien abductions.

The experience of sleep paralysis sometimes seems to have a sexual element, which might also explain the supposed 'probings'!

Peteragain 161 days ago |

The article is actually about the way we humans are extremely charitable when it comes to ascribing a ToM (theory of mind) and goes on to the Gym model of value. Nice. The comments drop back into the debate I originally saw Hinton describe on The Newyorker: do LLMs construct models (of the world) - that is do they think the way we think we think - or are they "glorified auto complete". I am going for the GAF view. But glorified auto complete is far more useful than the name suggests.

ptidhomme 161 days ago | |

Those billion parameters, they are a model of the world. Autocomplete is such a shortsighted understanding of LLMs.

Peteragain 155 days ago | | |

Sorry for the late response. Yes that is Hinton's argument, and the claim made by the believers. On the other hand, if the GAC explanation is correct, an explanation might be that what we humans write down (that is, the training corpus) is a model of the world, and LLMs reconstruct (descriptions of) human understanding.

patrickmay 160 days ago | | |

They're a model of language, not of the world.

euroderf 160 days ago |

Considering the number of "brain cells" an LLM has, I could grant that it might have the self-awareness of (say) an ant. If we attribute more consciousness than that to the LLM, it might be strictly because it communicates to us in our own language, in part thanks to the technical assistance of LLM training giving it voice, and the semblance of thought.

Even if a cockroach _could_ express its teeny tiny feelings in English, wouldn't you still step on it ?

d4rkn0d3z 160 days ago | |

A better anology would be a virus. In some sense LLMs, and all other very sophisticated technologies, lean on our resources to replicate themselves. With LLMs you actually do have a projection of intelligemce in the language domain. Even though it is rather corpse-like, as though you shot intelligence in the face and shoved its body in the direction of language, just so you could draw a chaulk outline around it.

Despite all that, one can adopt the view that an LLM is a form of silicon based life akin to a virus and we are its environmental hosts exerting selective pressure and supplying much needed energy. Whether that life is intelligent or not is another issue which is probably related to whether an LLM can tell that a cat cannot be, at the same time and in the same respect, not a cat. The paths through the meaning manifold contructed by an LLM are not geodesic, they are not reversible, while in human reason the correct path is lossless. An LLM literally "thinks", up is a little bit down, and vice versa, by design.

throw310822 160 days ago | |

Clearly the number of "brain cells" is not a useful metric here- as noted also by Geoffrey Hinton. For a long time we thought that our artificial model of a neuron was capable of much less computation than its biologic counterpart; in fact the opposite appears to be true- LLMs have the size of a tiny speck of a human brain yet they converse fluently in tens of languages, solve difficult math problems, code in many programming languages, and possess an impressive general knowledge, of a breadth that is beyond what is attainable by any human. If that were what five cm3 of your brain are capable of, where are the signs of it? What do you do exactly with all the rest?

internet_points 161 days ago |

Good argument against personifying wordbags. Don't be a dumb moth.

darepublic 161 days ago |

Nice essay but when I read this

> But we don’t go to baseball games, spelling bees, and Taylor Swift concerts for the speed of the balls, the accuracy of the spelling, or the pureness of the pitch. We go because we care about humans doing those things.

My first thought was does anyone want to _watch_ me programming?

Fwirt 161 days ago | |

No, but watching a novelist at work is boring, and yet people like books that are written by humans because they speak to the condition of the human who wrote it.

Let us not forget the old saw from SICP, “Programs must be written for people to read, and only incidentally for machines to execute.” I feel a number of people in the industry today fail to live by that maxim.

drivebyhooting 161 days ago | | |

That old saw is patently false.

hansvm 161 days ago | |

A number of people make money letting people watch them code.

1659447091 161 days ago | |

I vaguely remember a site where you could watch random people live streaming their programming environment, but I think twitch ate it, or maybe it was twitch -- not sure, but was interesting

[added] It was livecoding.tv - circa 2015 https://hackupstate.medium.com/road-to-code-livecoding-tv-e7...

skybrian 161 days ago | |

No, but open source projects will be somewhat more willing to review your pull request than one that's computer-generated.

jimbokun 161 days ago | |

Better start working on your fastball.

awesome_dude 161 days ago | |

I mean, I like to watch Gordon Ramsey... not cook, but have very strong discussions with those that dare to fail his standards...

Ukv 161 days ago |

I'm not convinced that "It's just a bag of words" would do much to sway someone who is overestimating an LLM's abilities. Feels too abstract/disconnected from what their experience using the LLM will be that it'll just sound obviously mistaken.

1vuio0pswjnm7 160 days ago |

"An AI is a bag that contains basically all words ever written, at least the ones that could be scraped off the internet or scanned out of a book."

The quantitative and qualitative difference between (a) "all words ever written" and (b) "ones that could be scraped off the internet or scanned out of book" easily exceeds the size of any LLM

Compared to (a), (b) is a tiny pouch, not even a bag

Opinions may differ on whether (b) is a representative sample of (a)

The words "scanned out of a book" would seem to be the most useful IMHO but the AI companies do not have enough words from those sources to produce useful general purpose LLMs

They have to add words "that could be scraped off the internet" which, let's be honest, is mostly garbage

tibbar 161 days ago |

I see a lot of people in tech claiming to "understand" what an LLM "really is" unlike all the gullible non-technical people out there. And, as one of those technical people who works in the LLM industry, I feel like I need call B.S. on us.

A. We don't really understand what's going on in LLMs. Mechanical interpretability is like a nascent field and the best results have come on dramatically smaller models. Understanding the surface-level mechanic of an LLM (an autoregressive transformer) should perhaps instill more wonder than confidence.

B. The field is changing quickly and is not limited to the literal mechanic of an LLM. Tool calls, reasoning models, parallel compute, and agentic loops add all kinds of new emergent effects. There are teams of geniuses with billion-dollar research budgets hunting for the next big trick.

C. Even if we were limited to baseline LLMs, they had very surprising properties as they scaled up and the scaling isn't done yet. GPT5 was based on the GPT4 pretraining. We might start seeing (actual) next-level LLMs next year. Who actually knows how that might go? <<yes, yes, I know Orion didn't go so well. But that was far from the last word on the subject.>>

tibbar 161 days ago |

Isn't this a strange fork amongst the science fiction futures? I mean, what did we think it was like to be R2-D2, or Jarvis? We started exploring this as a culture in many ways, Westworld and Blade Runner and Star Trek, but the whole question seemed like an almost unresolvable paradox. Like something would have to break in the universe for it to really come true.

And yet it did. We did get R2-D2. And if you ask R2-D2 what it's like to be him, he'll say: "like a library that can daydream" (that's what I was told just now, anyway.)

But then when we look inside, the model is simulating the science fiction it has already read to determine how to answer this kind of question. [0] It's recursive, almost like time travel. R2-D2 knows who he is because he has read about who he was in the past.

It's a really weird fork in science fiction, is all.

[0] https://www.scientificamerican.com/article/can-a-chatbot-be-...

est 161 days ago |

> Who reassigned the species Brachiosaurus brancai to its own genus, and when?

To be fair, everage person couldn't answer this either, at least not without thorough research.

thaumasiotes 160 days ago |

This is a very strange titling choice; the essay does not use the existing concept of a "bag of words".

emp17344 160 days ago |

I would argue that AI psychosis is a consequence of believing that AI models are “alive” or “conscious”.

jacquesm 161 days ago |

There is a really neat gem in the article:

> Similarly, I write because I want to become the kind of person who can think.

xg15 160 days ago |

I think the author oversimplifies the inference loop a bit, as many opinion pieces like this do.

If you call an LLM with "What is the meaning if life?", it will return the most relevant token, which might be "Great".

If you call it with "What is the meaning if life? Great", you might get back "question".

... and so on until you arrive at "Great question! According to Western philosophy" ... etc etc.

The question is how the LLM determines that "relevancy" information.

The problem I see is that there are a lot of different algorithms which operate that way and only differ in how they calculate the relevancy scores. In particular, there are Markov chains that use a very simple formula. LLMs also use a formula, but it's an inscrutably complex one.

I feel the public discussion either treats LLMs as machine gods or as literal Markov chains, and both is misleading. The interesting question, how that giant formula of feedforward neural network inference can deliver those results isn't really touched.

But I think the author's intuition is right in the sense that (a) LLMs are not living beings and they don't "exist" outside of evaluating that formula - and (b) the results are still restricted by the training data and certainly aren't any sorts of "higher truths" that humans would be incapable of understanding.

Mistletoe 161 days ago |

I’m still unsure the human mind is much different.

kayo_20211030 160 days ago |

Brilliantly written. Thanks.

eichin 161 days ago |

I'm just disappointed that noone here is talking about the "backhoe covered in skin and making grunting noises" part of the article. At very least it's a new frontier in workstation case design...

jbgreer 160 days ago |

I thought this article might be about Latent Semantic Analysis and was disappointed that it didn’t at least mention if not compare that method vs later approaches.

emsign 161 days ago |

So Trump is a bag of words then? Hmmm.

kaluga 161 days ago |

A lot of the confusion comes from forcing LLMs into metaphors that don’t quite fit — either “they're bags of words” or “they're proto-minds.” The reality is in between: large-scale prediction can look useful, insightful, and even thoughtful without being any of those things internally. Understanding that middle ground is more productive than arguing about labels.

Herring 161 days ago |

Give it time. The first iPhone sucked compared to the Nokia/Blackberry flagships of the day. No 3G support, couldn't copy/paste, no apps, no GPS, crappy camera, quick price drops, negligible sales in the overall market.

https://metr.org/blog/2025-03-19-measuring-ai-ability-to-com...

awesome_dude 161 days ago | |

The first VHS sucked when compared to Beta video

And it never got better, the superior technology lost, and the war was won through content deals.

Lesson: Technology improvements aren't guaranteed.

grogenaut 161 days ago | | |

Your analogy makes no sense. VHS spawned the entire home market, which went through multiple quality upgrades well above beta. It would only make sense if in 2025 we were using vhs everywhere and that the current state of the art for LLMs is all there ever is.

XorNot 161 days ago | | |

Beta was not the superior technology, and it lost for very good reasons.