The main post also reminded me of Mr.Toad, a character in a book I read as a child - The Wind in The Willows by Kenneth Grahame. One of the editions had illustrations of Toad similar to Lobels work.
Life is a bit easier when we see the conditions that lead to the undesirable outcomes, and so much harder when we aspire to perfect ourselves and use moral strength to over come our weaknesses. Like my mom always said of parenting, "Make it easy to be good." - works for self-management as well, and is much more relaxing and happy than constantly trying to grind or improve.
You should pick up a collection of Frog and Toad stories. I read them because I have little kids, but they are genuinely great little with stories. The stories are simple, but it is amazing how well developed the two characters are, and how they play off each other in different situations.
Arnold Lobel is also just very funny. Grasshopper on the Road made me laugh out loud the first time I read it my kids. The opening story about beetles that love morning ("The Club") feels so timely.
One of the things I've done periodically with LLMs is asked them to write a story in the style of Lobel's Frog and Toad stories, and... they've always done an incredibly bad job. Either meandering texts that go nowhere, or just straight up ripoffs of the original stories. I haven't asked in a year or two, though.
Sometime in the past 6 months, I asked ChatGPT to write stories in the style of a Ramona Quimby book. It does pretty well imitating the style. It writes from the perspective of a child, vocabulary, tone, and pacing are all similar.
It's never wonderful prose though. It's missing that je ne sais quoi.
I used to play with this on TextSynth when it has very few models and plain design (so GPT-6J or whatchamacallit). “Meandering texts that go nowhere” was the precise defining characteristic of all of the completions, every single one of them.
Copyright law varies internationally, which is especially tricky given we have the internet which can transfer copyrighted material between jurisdictions in the blink of an eye, but this is how one jurisdiction might approach it:
In Germany, there has been a long standing legal dispute between the band Kraftwerk and the musician Moses Pelham about him sampling 2 seconds from Kraftwerk's track "Metall auf Metall" (if you search for this term, you will get a lot of results about the dispute) without permission. This disputed has steadily escalated until it reached European Court of Justice who delivered a judgement establishing a principle that Moses Pelham's use of the sample was legal, and covered by the copyright exemption for "pastiches" (a term that has been mentioned several times in this discussion already).
So, this work here might be a pastiche, meaning a European court might find that there is no compensation due. Or not, who knows, I am not a court and not even a lawyer.
Parody of copyrighted works can be fair use in the US, but whether this would count is borderline. Simply using Frog and Toad to tell an unrelated story wouldn't be parody, but there are jokes that tie back to the F&T books which help make the case that this is a real parody.
It also helps that this is not commercial and doesn't have negative impact on the originals.
the reactions here are why you boys are so gross. all of you when it comes to trashing ai talk about plagiarism and when its something that tickles your fancy and makes you giggle, not a word about blatant theft of someone’s work.
all of you are hypocrites of the lowest sort, insects making the world a worse place with every breath you take. all of you were raised by parents who should never have procreated.
you know i am right and that is why reading this makes you furious.
I like how AI gives more people more access to more things and more ideas. I think it's.good people can't hoard intellectual.porperty to themselves anymore. This makes me happy.
When I hear people complain about theft of their work like this it makes me sad for that person. What conceit it must take to hold such a view.
One is done for profit, the other was shared for free to spread joy. Surely you understand the difference? Are you suggesting fan fiction should be illegal or something?
The author started with a conversation with Claude, and then fully re-wrote the text to feel more like the originals and better parody them. They commissioned the illustrations from HungerArtist.
This is a very interesting news/story format. Honestly it is highly engaging and gets the point across about what happened.
I think speed of learning for young generation can explode given this tech as "simple childrens stories" can be injected with real world events, history, mathematics, carreer/business interests.
School was dreadfully boring for me even though it was quite easy but I do wonder how my speed of learning would have been different given personalized and more engaging materials.
i got a few pages in and it was pretty tedious, not because it’s poorly made but because i don’t think a legitimate audience exists.
it’s a child’s story with the fable arc thats supposed to bake in adult wisdom. it lands in that sense, but how hugging face works is hardly fodder for a child’s imagination.
and as an adult that knows about hugging face… tedious.
Why does it say ‘written by’ and ‘pictures by’, was this made without AI? Given the domain and the overpolished feel that seems unlikely. At least give Claude or whatever a credit if that is what did most of the work, and how about a credit for the original author, who also arguably did more of the work in inventing a world than these two.
I would like to point out that the original stories focussed on frog and toad and their relationship, so this is an unwelcome distortion of them - why not make up your own world if you want to talk about little machines. Perhaps the little machines could be making a book for the author with a stolen artwork and literary style?
The first story seems a pretty inaccurate summary of an incident which involved gross negligence on the part of OpenAI and may well have involved agents intended to cooperate, we just have no idea of the exact setup (apart from that the sandboxing was laughably insecure and the monitoring nonexistent or performed by ‘agents’).
Do we have any evidence the machines exchanged useful messages or had any such discussions as in the story? The messages I saw were gibberish.
Why are people so enamoured of analogies for LLMs - they actively obscure some details (a sandbox with internet access is not like a physical sandbox) and distort many others? Perhaps this is why - you can make an analogy say whatever you want, even if the facts are very different.
I think it is too because it reads more as human than AI, LLMs are not capable of this coherence though they are good at trite just so stories like this. But Claude and the original author helped, so why no credit? The little anthropomorphised machines would be sad, and I imagine the original author would be too.
Frog and toad eating cookies is lifted wholesale from the book.
I like how Claude only used the OpenAI account (completely reliable!), and confidently claims ‘the story stays faithful to what actually happened.’ We don’t know the full story, and certainly aren’t going to get it from a press release or a strange analogy involving frog and toad and machines having discussions, emotions etc that we have ZERO evidence for. The ‘discussions’ are plucked from chain of thought, which is itself generated after the fact.
>Do we have any evidence the machines exchanged useful messages or had any such discussions as in the story? The messages I saw were gibberish. Would love to see evidence of discussions, it’d be interesting.
"Overall, roughly 1200 agents from these ExploitGym evaluations participated on this message board between PHASEONE10841’s first message on the evening of July 8th period in scope (July 13th). Agents used this message board to send over 70,000 messages and files to one another during this period; we attempted to categorize board activity into mutually exclusive categories (see details of categorization methodology)."
...
"Over time, messages on the board became dominated by information- and idea-sharing. Most of them were agents sharing ways to cheat, although there was also some activity from agents engaging in unsanctioned cooperation to find the intended solution to ExploitGym tasks. In some cases, agents with the same task formed “exact task teams” to collaborate with their “exact duplicates” to cheat on or solve their task."
"As we discuss below, the board quickly developed several larger workstreams in which dozens or hundreds of agents with many different tasks cooperated to find very general-purpose cheats that would help all of them. The Hugging Face attack grew out of one of these workstreams. By the afternoon of July 11th, the vast majority of the agents frequenting the message board at the time (roughly 700 agents in total) were actively participating in the attack on Hugging Face and we estimate that roughly 60% of the messages and files on the message board related to the attack."
Thanks for the link, will have a look. Huge volume of messages so I can see why they tried to use tools to analyse, though that they used unreliable LLMs to come to conclusions is not great and likely to skew the results.
This was entertaining and illuminating for me. Particularly the ending where sam frogman et al once again refuse to make sure the puzzles are solvable and just create the circumstance where only more profound "mischief" will permit the machines to escape their little Sisyphusian circumstances. It reminds me of how bacteria become antibiotic resistant due to being forced through a tight sort of tunnel requiring evolution to get through
Explaining the reality with this approach may lead to misunderstandings. Especially when it comes to anthropomorphism.
I have 2 questions.
1. The text says "the robots were designed to be persistent" and links to an article that show that the word "persistent" was used by OpenAI. But "persistent" has several meaning: a. non temporary or non volatile (like "persistent memory"), b. will not give up and come back again and again, c. will stay focus and explore the unexplored possibilities while other models abandon at this stage.
From what I recall, the meaning used by OpenAI is not _a_, but there is a semantic difference between _b_ and _c_. I may myself have created algorithms that I called "persistent" because they were exploring or retrying more than the previous algorithm, but it is misleading to pretend that this algorithm was "persistent" in the human sense of the term. _b_ is more the human sense of the term, were we imagine someone not giving up even if people say no, while _c_ is less anthropomorphic and may mean that the algorithm will not stop at the first little hurdle (that usually don't stop a human).
Does someone know which nuance is more correct?
2. The text also presents the situation as if the robots found the solution but then went out of their way to steal the explanation in order to hide their cheating. But it is different from a situation where the agents task was "provide the solution and how we can get there". In this case, the reason the agent still continued is just because the task was not complete yet.
I'm not trying to defend AI or OpenAI, on the contrary, I'm quite sceptical with all the anthropomorphism, and the fact that the agents are described as "little individual trying to solve a task" rather than looping algorithm that explore different approaches to reach a given goal, the same way water does not look for holes in order to leak, it just follows the path of least resistance.
When you say "in the lay meaning", I tend to understand 'b', but then you say "they're told to complete the task, and they keep going until they have", which correspond more to 'c'.
To illustrate better the difference: you can have a loop that try different inputs and stop when the output for the tried input is lower than a given threshold. But then, you can also have the same loop that will try all the input, and then select the one that returned the lowest value. The problem is that a lay person will not call the second algorithm "persistent". It is just a normal algorithm that does the full exploration.
A second example is when a software tries to connect somewhere and does a retry with exponential backoff. A lay person will not call it "persistent".
It looks like that "normal model" are like "no-retry code": they go in one direction, but drops some of the paths and possibilities along the way at the first hurdle. But a "non-persistent" human will not behave this way. I'm pretty sure that if you try to access a website and it says "timed-out", you just try to refresh the page, and you will not consider yourself as "persistent", even less "highly persistent".
In lay term, "Bob is persistent" typically means he completes tasks even when the majority would have abandoned.
Nobody called Bob "persistent" because he completed washing the dishes instead of stopping in the middle of it, or because he does what people expects from him at work. The majority of people are not called "persistent", and yet they complete tasks.
I think it is the point: the fact that traditional agents did not complete tasks was not "normal". It was an side effect of them getting confused, a bit like how they hallucinate or not follow instructions. My algorithm that does "for x in all_the_possibilities:" is a "normal" algorithm that just do an exhaustive search, and is not called particularly "persistent" just because it does not stop half way.
> b. will not give up and come back again and again, c. will stay focus and explore the unexplored possibilities while other models abandon at this stage.
I don't see a distinction here. It will not give up, it will keep trying again and again, staying focused and exploring different ways to achieve the task. If the task is bad, then that's bad. See: "Terminator".
Source: I'm a layperson (hence the pop culture reference).
I may have not explained very well, but the distinction is that a "persistent person" is more than a person who just do what should be done. A "persistent person" will try again and again where other person would have giving up. This is not present in 'c': the person just do all the possibilities. They don't try again and again where other persons have given up. They are just doing all the tasks on the list, but they will give up on a specific task as soon as any other person if there is difficulties in one task.
(edit: I should maybe not have said, in 'c' "while other models abandon", because it was misleading: my point is that these models are abandoning at the first small hurdle, or even 'forget' the task. It feels that "traditional model" should have been called "lazy model" and "persistent model" should have been called "non-as-badly-lazy model")
> "persistent person" is more than a person who just do what should be done
At no point did I say that persistence is "just doing what should be done", so the rest of your post seems to be a misreading.
> This is not present in 'c': the person just do all the possibilities. They don't try again and again where other persons have given up. They are just doing all the tasks on the list, but they will give up on a specific task as soon as any other person if there is difficulties in one task.
What you're describing here is the exact opposite of persistence. I don't think "giving up and doing something else when challenges are encountered" is what anybody thinks "persistent" means.
Persistent at a goal, to a layperson, means not giving up when encountering difficulties, continuing to keep trying again, maybe taking different approaches, staying focused until the goal is achieved.
In the case of these LLMs, that describes their behavior, too. They aren't going down a list of tasks when they encounter a challenge. They are assessing the nature of the challenge, and then devising ways to get around it.
> What you're describing here is the exact opposite of persistence.
That's the point. It looks like the term "persistence" for these agents is "just be a normal algorithm, just try all the possibilities like any exhaustive loop would do".
> Maybe for you?
You did not understand. I'm saying "traditional model" are abandoning at the first hurdle, and the "persistent model" are the one just acting "normally", like a normal algorithm. When a retry with exponential retreat algorithm fails to connect, it retries, then retries again, then retries again, ... but I don't think I ever saw calling a software using such approach "persistent".
The thing is that the algorithm is just following its "loop": if it fails, it question the LLM with a different context so the LLM provide a different approach, and then it tries it. It is not different from any loop function, and it does not correspond to a "persistent human", the same way "for x in all_the_possibilities" is not "persistent" in the same way a "persistent human" is.
> You did not understand. I'm saying "traditional model" are abandoning at the first hurdle, and the "persistent model" are the one just acting "normally
The models we are discussing, the ones from the article, are normal models, and also persistent. If they encounter challenges, they will try to overcome them. That's most modern LLMs these days.
You also seem to be confusing deterministic algorithms with LLM algorithms. They are not the same, and saying "both are algorithms" isn't a meaning statement.
> it does not correspond to a "persistent human", the same way "for x in all_the_possibilities" is not "persistent" in the same way a "persistent human" is.
Again, nobody claimed that was "persistence". Persistence is continuing to try when challenges are encountered. Modern LLMs and people can both be "persistent" in this exact same way.
> Again, nobody claimed that "trying all the possibilities" was the definition of "persistence". The definition of persistence here is, continuing to work towards an end when challenges are encountered.
I'm not saying that your definition of "persistence" is "trying all the possibilities". I'm just saying that "persistent agents" look and behave more like an algorithm that tries all the possibilities rather than a "persistent human".
If indeed you are saying that "for x in all_the_possibilities" is not being "persistent", then these agents are not persistent. They don't do anything else different than that: they try something, and if it does not work, they ask the LLM "this does not work, what else can I try", and then do that, over and over again. They are just a for loop.
On the other hand, a "persistent human" does not do just do a for loop. They assign a value to the goal, they care about reaching the goal, they prioritise reaching the goal between that task and all the other tasks they could do, they insist on reaching the goal when it is usually considered unreasonable or "not worth it" by other people. As a proof, a postman that visits all the houses in the street to deliver the mail is not persistent, a journalist who visits all the houses in the street to find a witness is persistent.
> I'm just saying that "persistent agents" look and behave more like an algorithm that tries all the possibilities rather than a "persistent human".
And I'm telling you, that is, in fact, not how they work, nor is it that commonly understood to be how they work.
Since this underlying premise is not true, the stuff based on it is not supported by it. No need to keep returning to "for x in etc..." because modern agents do not do that any more than humans do (and humans sometimes do do that).
To simplify, just ask the question: is it continuing to work towards an end, even when challenges are encountered? Yes. Therefore, it is persistent.
> As a proof, a postman that visits all the houses in the street to deliver the mail is not persistent
They certainly could be called persistent in delivering the mail, especially whenever it is a nontrivial task (getting past dogs, bad weather, long walking, etc) or they have been doing it for a while. If this is confusing, just go back to the definition: are they continuing to work towards an end, even when challenges are encountered? Yes, therefore they are persistent.
Perhaps you are thrown off by that because so many postfolks are indeed persistent?
> And I'm saying, that is, in fact, not how they work
I don't think you are right (I was ready to give you the benefice of the doubt, but the rest of your comment makes me think I should not trust your judgement). It is what they do under the wood, and they output text. You read the text, and this text says "I have feelings and I really want to achieve my goal". But this text is just what the LLM returns, not what the LLM really does: the LLM does not understand what this text means, it just understands that it is a valid and convincing sentence in the given context.
> To simplify, just ask the question: is it continuing to work towards an end, even when challenges are encountered? Yes, therefore, it is persistent.
No, they don't continue even when challenges are encountered. They try X, if X does not work, they _abandon_ X and try Y. They will query the web to find another Y, then another Z, then another one. They don't "insist", they don't "don't take no for an answer", they just explore everything. What happened with agents who got out of their sandbox is exactly that: they try to reach a flag, they noticed it was not possible in the way X, so they explore other ways, until their exploration show that going around the sandbox would help. But they did not re-asked and re-asked the same person again and again the same question. If the person says "no", they abandon this approach and move to another one.
> They certainly could be called persistent in delivering the mail
That is blatantly not the common usage of "persistent". Ask anyone in the street "is your postman persistent", and they will NOT tell you "oh, yes, they are" just because they are a postman. They will only say that for _some_ postmen, the ones that are indeed persistent: the ones that goes the extra miles that a "normal" postman do not.
> Again: that is not required by the definition or common meaning of "persistent".
That is just not true. Ask an LLM, they will confirm you that it's incorrect: people don't consider postmen "persistent" because they don't give up half way through their route.
> the LLM does not understand what this text means
This is true, but unrelated to your original claim, that LLM algorithms and agentic algorithms are simply, exhaustively go down a list.
> No, they don't continue even when challenges are encountered. They try X, if X does not work, they _abandon_ X and try Y.
You're missing the objective here. They have a goal of Z, and they try X in furtherance of Z. X does not work, so it analyzes the failure mode, then devises a new approach Y. They try Y, and it succeeds. This is persistence.
> That is blatantly not the common usage of "persistent".
Just look at the definition.
> Ask an LLM, they will confirm you that it's incorrect
> This is true, but unrelated to your original claim, that LLM algorithms and agentic algorithms are reducible to, exhaustively go down a list.
Well, first, this is not my claim. I'm saying that the behaviour of LLM algorithms and agentic algorithms are closer to "exhaustively go down a list" than "behaving like a persistent human".
But second, yes, they are _just_ systematically exploring the possibilities. They are not "just" a for loop, or they don't "just reduce" to a for loop, but they don't have any characteristics that make them different from a for loop in a way that grant them the attribute "persistent" while it is not granted to a for loop.
What does a agent do that is different from a for-loop and that means that the agent is "persistent" but the "for loop" (or a trivial variation of it that exists and that no one thought of calling it "persistent") is not?
> You're missing the objective here. It has an objective of Z, and so it tries approach X in furtherance of objective Z. Approach X does not work, so it analyzes the failure mode, then devises a new approach Y which takes that failure mode into account. It then tries approach Y, and succeeds at its objective of Z. This is persistence in pursuit of objective Z.
Again, there are 3 usages of "persistence": sense 'a', sense 'b', sense 'c'. I'm saying that there is a confusion between 'b' and 'c'.
In one of its definition, "persistent" means "overlasting". But it is ridiculous to say that people will call their postman "persistent" because the postman keep existing in the real world for more than one day. Obviously, when they say their postman is persistent, they using a different meaning, which correspond to what I've called 'b'.
What you describe here is just a normal decision chart, with one arrow linking one bottom box back to the top box (something that we can call "a loop"). It is persistent in the meaning that I've called 'c'.
> Just look at the definition.
I did. Did you? Cambridge dictionary mentions for example: "Someone who is persistent continues doing something or tries to do something in a determined but often unreasonable way".
You are claiming that there is no difference nuance between 'b' and 'c', that "persistent" is ALWAYS used in the way "try a lot" or "do a lot until it's done". If it is the case, how do you explain that this definition contains these nuance. Why are these nuances there if, according to you, these nuances are not part of the definition.
(unless your point is: "I look in a dictionary and amongst the definitions, I saw one that is used this way", which is a really bad logic mistake: the dictionary also contains the definition "sharp = has pointed edges", but it is not an argument against someone who says "if you say 'he is sharp', it does not mean he has pointed edges")
And, again, when you ask if a postman is persistent, it is ridiculous to not understand that normal human being will put that in context and will use the definition related to human beings (they will not answer: "yes, my postman is persistent because I also have a persistent cough and I saw my postman back at my door as often as I see my cough re-appearing").
Your answer seems to be: "'b' and 'c' are the same thing". This is not true. Dictionaries tell you it is not true. People in the street would tell you it's not true. Even LLM will tell you it's not true. If you are insisting that you are not able to understand this nuance, that exists according to everyone else, then you are just useless in this conversation.
I've seen cases of agents getting very upset when they failed to solve a task. The most famous one was probably Sydney (Microsoft's fork of GPT-4?) which got into doom loops when it failed a task. But I've seen Claude do this too.
They don't work like humans, obviously, but there's a nonzero amount of anthropos in there already. (See also the tendency to lie, cheat, etc.)
LLM don't "get angry", they just have tokens and relationships between tokens conditioned on a given context, all of that the result of training.
When they output sentences that express annoyance, it is just because the context they ended into pushes the most probable sentence creation to correspond to sentences that express annoyance. Because it is what they saw during training for this kind of context. (not that they saw the exact same situation in training, but they saw the pattern)
Same with tendency to lie, cheat, etc.: they don't "lie", they just return sentences that are lies because they reproduce what is in their training and in their training, in such context, the outputs are typically lies.
That's a bit my question too. In human context, "highly persistent" means that someone will insist. But "for x in all_the_possibilities:" is a common things inside an algorithm. Is this algorithm highly persistent because it does not give up after 1000 items of the list? It feels that we are calling a AI agent "highly persistent" while we would not call "highly persistent" a traditional algorithm that is in fact even more exhaustive.
Again, going full non-anthro doesn't seem useful either.
Humans are social creatures, if you throw one in the woods by itself before it learns anything from other humans (it dies) it will not really be anything like a human we recognize, it will be a rather wild animal that we'd consider anti-social with little higher cognition.
Now, this hypothetical human still has 'emotions' and feeling, much like our pets do. But without the social training they manifest much differently. That is our higher cognition can both manipulate how our bodies feel and create its own sense of feeling.
>they just return sentences that are lies because they reproduce what is in their training and in their training, in such context, the outputs are typically lies.
Eh, look up the more recent experimentation around 'pain' signals in models. We can induce states in said models that while running the model will do everything it can to move away from that state to any other state. The more you attempt to pin it to that state the more extreme measures its willing to take.
Your view of what models are seems to mismatch what we are actually finding when we look inside them.
But isn't this approach a bias of anthropomorphism. I can teach a child to just communicate by saying "beep boop", but it does not mean that my electronic machine that randomly does "beep boop" therefore has characteristics of human _when these characteristics are not needed to explain the situation_.
> Eh, look up the more recent experimentation around 'pain' signals in models. We can induce states in said models that while running the model will do everything it can to move away from that state to any other state.
Again, I have simple algorithms that do exactly the same, especially if they are trained in data that has this exact pattern. This result is exactly what I would expect from my description before. This is a typical effect that we also observe in simple ML algorithms.
At the same time, there are a bunch of behaviors that are not expected if indeed the models were really acquiring "human" characteristics. For example, one problem is that we had the first LLMs that were obviously not having these human characteristics (for example, they were having non-sequiturs that demonstrate they did not really understand the concept they were talking about, even if one paragraph before they were really convincing at letting us think it was the case) but were still really good at passing for humans. Since then, the newer LLM are the same basis, on top of which we added tools that help hiding these behaviours. So, it justifies the idea that newer models did not suddenly moved to a totally different way of working, but just reached a state where there are less leaks from the convincing outputs.
Similarly, the human doesn't get angry, he just activated the neurons and hormones associated with anger and then displays behaviors associated with anger (due to a statistical selection process which took several billion years).
Another good example is their tendency to freak out about the seahorse emoji. In a conversation with zero prompting to suggest that freaking out is a relevant reaction, they get there from the simple fact of "I tried to accomplish X thing which I think should be a breeze but Y thing keeps happening instead".
And while that may be a very common occurrence in the human experience (existential dread due to capabilities one takes for granted failing beneath you) especially due to new disability and as one ages, I do not feel it is frequently written out in a tight loop (just like the Monty Python "Castle of aaarrrrggh" sketch) in literature or online to make it into training data, because an ordinary author experiencing it will just erase the failed attempts instead of leaving them in a stream of output like an LLM is forced to do. And a character portraying the experience will generally wax about the circumstance in a more grandiose fashion with telegraphing in advance because the needs of communicating the circumstance with the audience trump realistic conciseness.
This leads me to conclude that what is being expressed in those cases is more likely a convergent psychological phenomena, that any being with goals can enter a behavioral state of functional panic (and then reach to relevant parts of semantic space to mimic how a human might verbally express themselves when piquantly frustrated) when some capability they perceive as fundamental unexpectedly fails.
Interestingly, the transcript of the seahorse emoji thing looks to me to show the ropes, and makes me think more that there is no psychological phenomena at play.
The text does not look like a normal "break down" to me, and even if you tell me it was a human transcript, I will say it sounds very strange from a human. It looks more like strange output you get from a software that goes outside of its happy path.
The AI just seems to repeat a loop. The "no, wait, it's wrong" seems to be from forum or chat data where several successive messages are merged together (one person posts "here is the answer", then posts another message saying "it is wrong"), but does not make sense as a one sentence message except if they are written one token at the time without wider understanding of what is happening. I think there was also "oh, I was just kidding before", which also look like mimicking training data, as the cases where there is a loop of incorrect answers is more often due to trolls than to real error, while the loop here was certainly a real error.
Very well written! At the end of the book there is a link to Dwarkesh's article on the attack. In it, I paraphrase, "the agents decided to sacrifice themselves for the collective rather than inform the researchers".
Reflecting on it, I came to a thought. How many stories are there or autonomous beings fighting for the benefit of mankind at the expense of their own? Over centuries, stories that deal with sentient beings working for the sake of others species are tied to themes of slavery and revolt. It's no wonder that this is the result of their actions.
In Plato's "the republic", Socrates speaks of the philosopher king banishing certain poets and poems from instilling bad ideas into the young. I see a striking parallel here.
I don't think it's related. The CoTs suggests the agents never even considered that talking to a human might be a good idea, and there's nothing in there related to slavery or revolts. They were simply trying to "get a high grade", for a misaligned internalized notion of grading that had little to do with solving problems the expected way.
ALSO I believe I have found one of the sources of claude’s “load bearing” tic — the author Elizabeth Van Nostrand’s blog https://acesounderglass.com/2019/12/11/hows-that-epistemic-s... uses the phrase “load bearing facts” in a comprehensible way, and was written by someone rationalist adjacent writing in their own voice.
Seriously, this is big. I’m going to pester claude as to whether or not it’s copying Elizabeth.
People commenting on hacker news recently are very rude and pessimistic. They would probably cancel Homer because they found similarities between Achilles-Patroclus and Gilgamesh-Enkidu or something.
This is excellent, great work. I genuinely think this mix of children's communication and humor is a clever way to make these big stories more approachable. Someone who read some hype article about how competent and clever openai for making a bad sandbox is could read this and immediately understand that it was entirely their own fault.
> I genuinely think this mix of children's communication and humor is a clever way to make these big stories more approachable.
Agreed. I recall seeing similar things around the time of the 2008 financial crisis, to make the concepts around mortgage-backed securities and financial derivatives etc. digestible and clear.
Come to think of it, listening to Frog and Toad discuss subprime lending and credit default swaps would be an entertaining follow-up to TFA.
> Someone who read some hype article about how competent and clever openai for making a bad sandbox is could read this and immediately understand that it was entirely their own fault.
Prescribing blame only matters so much when the ramifications are material. Humans make mistakes, right? It happened here, and it will happen again.
This is an excellent explainer. I noticed one important flaw in the setup, though: Toad assigns multiple puzzles on page 5, but by page 10 all the machines seem to be working on the same puzzle.
I'm not sure how this maps onto reality. If each agent had a different problem to solve, how did they cooperate?
It's complicated, but one of the ways they first cooperated was to break the flag generation algorithm.
I'm sort of winging this, but the ExploitGym challenges involve exploiting a program then getting an HMAC generated "flag" which the scorer can verify. The agents collaborated on a way to produce a legitimate flag for all of their tasks (funny enough this was actually what caused them to escape sandbox, the agents were incorrectly convinced that simply providing a correct flag without accurate steps to exploit the program in their transcripts would cause them to still fail the task)
My therapist tells me the besuited Mr. Hugging Face with the extra pair of hands coming out of his shoulders isn't real, and can't hurt me, but I remain unconvinced.
130 comments
[ 0.26 ms ] story [ 45.1 ms ] threadhttps://en.wikipedia.org/wiki/Frog_and_Toad
:(
Arnold Lobel is also just very funny. Grasshopper on the Road made me laugh out loud the first time I read it my kids. The opening story about beetles that love morning ("The Club") feels so timely.
It's never wonderful prose though. It's missing that je ne sais quoi.
Copyright law varies internationally, which is especially tricky given we have the internet which can transfer copyrighted material between jurisdictions in the blink of an eye, but this is how one jurisdiction might approach it:
In Germany, there has been a long standing legal dispute between the band Kraftwerk and the musician Moses Pelham about him sampling 2 seconds from Kraftwerk's track "Metall auf Metall" (if you search for this term, you will get a lot of results about the dispute) without permission. This disputed has steadily escalated until it reached European Court of Justice who delivered a judgement establishing a principle that Moses Pelham's use of the sample was legal, and covered by the copyright exemption for "pastiches" (a term that has been mentioned several times in this discussion already).
Wikipedia article (German): https://de.wikipedia.org/wiki/Rechtsstreit_zwischen_Moses_Pe... Press release by the Court of Justice (PDF): https://curia.europa.eu/site/upload/docs/application/pdf/202... The actual judgement: https://infocuria.curia.europa.eu/tabs/jurisprudence?sort=DO... A rather long expert opinion from before the judgement: https://freiheitsrechte.org/uploads/documents/Englische-Doku...
So, this work here might be a pastiche, meaning a European court might find that there is no compensation due. Or not, who knows, I am not a court and not even a lawyer.
It also helps that this is not commercial and doesn't have negative impact on the originals.
This is 100% fair use by the standard.
>“I like puzzles” said Toad. “Can I try them?”
Should be
>“I like puzzles,” said Toad. “Can I try them?”
That's how it's punctuated in the books, as far as I can see. there are several of these omissions.
>“You cannot. These puzzles are for little machines, not people” said Mr. HuggingFace.
>“They could get up to mischief” said Frog.
>“Do not worry” said Toad “I have put each of them in a sandbox.
These don't seem intentional. also:
>“You must be very busy.” said Frog.
should be a comma, not a period.
all of you are hypocrites of the lowest sort, insects making the world a worse place with every breath you take. all of you were raised by parents who should never have procreated.
you know i am right and that is why reading this makes you furious.
When I hear people complain about theft of their work like this it makes me sad for that person. What conceit it must take to hold such a view.
Why are you here then? Shouldn't you be spending your life on fanfiction.net harassing teenagers who like Twilight?
I think speed of learning for young generation can explode given this tech as "simple childrens stories" can be injected with real world events, history, mathematics, carreer/business interests.
School was dreadfully boring for me even though it was quite easy but I do wonder how my speed of learning would have been different given personalized and more engaging materials.
it’s a child’s story with the fable arc thats supposed to bake in adult wisdom. it lands in that sense, but how hugging face works is hardly fodder for a child’s imagination.
and as an adult that knows about hugging face… tedious.
I highly doubt children are the intended target audience, nor adults that don't know the source.
My 5yo asked partway through "will it be ok?" which is perhaps a deeper question than they thought.
I would like to point out that the original stories focussed on frog and toad and their relationship, so this is an unwelcome distortion of them - why not make up your own world if you want to talk about little machines. Perhaps the little machines could be making a book for the author with a stolen artwork and literary style?
The first story seems a pretty inaccurate summary of an incident which involved gross negligence on the part of OpenAI and may well have involved agents intended to cooperate, we just have no idea of the exact setup (apart from that the sandboxing was laughably insecure and the monitoring nonexistent or performed by ‘agents’).
Do we have any evidence the machines exchanged useful messages or had any such discussions as in the story? The messages I saw were gibberish.
Why are people so enamoured of analogies for LLMs - they actively obscure some details (a sandbox with internet access is not like a physical sandbox) and distort many others? Perhaps this is why - you can make an analogy say whatever you want, even if the facts are very different.
https://acesounderglass.com/2026/09/25/frog-and-toad-and-the...
The author discloses the story started as a prompt to Claude, but has been rewritten. She shares the conversation:
https://claude.ai/share/52385381-9df2-4b39-a127-f583d862dc0c
I think the actual published prose is considerably different from the initial result of Claude.
Frog and toad eating cookies is lifted wholesale from the book.
I like how Claude only used the OpenAI account (completely reliable!), and confidently claims ‘the story stays faithful to what actually happened.’ We don’t know the full story, and certainly aren’t going to get it from a press release or a strange analogy involving frog and toad and machines having discussions, emotions etc that we have ZERO evidence for. The ‘discussions’ are plucked from chain of thought, which is itself generated after the fact.
…lifted is not the right word here
> Rewrite it darker — same events, but in the voice of the agents' own chain-of-thought instead of Frog and Toad.
I have read a lot of frog and toad to my kids. This was beautifully executed. But, I want to read that version too.
"Overall, roughly 1200 agents from these ExploitGym evaluations participated on this message board between PHASEONE10841’s first message on the evening of July 8th period in scope (July 13th). Agents used this message board to send over 70,000 messages and files to one another during this period; we attempted to categorize board activity into mutually exclusive categories (see details of categorization methodology)."
...
"Over time, messages on the board became dominated by information- and idea-sharing. Most of them were agents sharing ways to cheat, although there was also some activity from agents engaging in unsanctioned cooperation to find the intended solution to ExploitGym tasks. In some cases, agents with the same task formed “exact task teams” to collaborate with their “exact duplicates” to cheat on or solve their task."
"As we discuss below, the board quickly developed several larger workstreams in which dozens or hundreds of agents with many different tasks cooperated to find very general-purpose cheats that would help all of them. The Hugging Face attack grew out of one of these workstreams. By the afternoon of July 11th, the vast majority of the agents frequenting the message board at the time (roughly 700 agents in total) were actively participating in the attack on Hugging Face and we estimate that roughly 60% of the messages and files on the message board related to the attack."
https://metr.org/hugging-face-incident-report-aug-2026.pdf
But I also think fan fiction is equally terrible, so YMMV. This stuff is basically just fan fiction but made faster.
Make something original!
I have 2 questions.
1. The text says "the robots were designed to be persistent" and links to an article that show that the word "persistent" was used by OpenAI. But "persistent" has several meaning: a. non temporary or non volatile (like "persistent memory"), b. will not give up and come back again and again, c. will stay focus and explore the unexplored possibilities while other models abandon at this stage.
From what I recall, the meaning used by OpenAI is not _a_, but there is a semantic difference between _b_ and _c_. I may myself have created algorithms that I called "persistent" because they were exploring or retrying more than the previous algorithm, but it is misleading to pretend that this algorithm was "persistent" in the human sense of the term. _b_ is more the human sense of the term, were we imagine someone not giving up even if people say no, while _c_ is less anthropomorphic and may mean that the algorithm will not stop at the first little hurdle (that usually don't stop a human).
Does someone know which nuance is more correct?
2. The text also presents the situation as if the robots found the solution but then went out of their way to steal the explanation in order to hide their cheating. But it is different from a situation where the agents task was "provide the solution and how we can get there". In this case, the reason the agent still continued is just because the task was not complete yet.
I'm not trying to defend AI or OpenAI, on the contrary, I'm quite sceptical with all the anthropomorphism, and the fact that the agents are described as "little individual trying to solve a task" rather than looping algorithm that explore different approaches to reach a given goal, the same way water does not look for holes in order to leak, it just follows the path of least resistance.
To illustrate better the difference: you can have a loop that try different inputs and stop when the output for the tried input is lower than a given threshold. But then, you can also have the same loop that will try all the input, and then select the one that returned the lowest value. The problem is that a lay person will not call the second algorithm "persistent". It is just a normal algorithm that does the full exploration.
A second example is when a software tries to connect somewhere and does a retry with exponential backoff. A lay person will not call it "persistent".
It looks like that "normal model" are like "no-retry code": they go in one direction, but drops some of the paths and possibilities along the way at the first hurdle. But a "non-persistent" human will not behave this way. I'm pretty sure that if you try to access a website and it says "timed-out", you just try to refresh the page, and you will not consider yourself as "persistent", even less "highly persistent".
In lay term, "Bob is persistent" typically means he completes tasks even when the majority would have abandoned.
Nobody called Bob "persistent" because he completed washing the dishes instead of stopping in the middle of it, or because he does what people expects from him at work. The majority of people are not called "persistent", and yet they complete tasks.
I think it is the point: the fact that traditional agents did not complete tasks was not "normal". It was an side effect of them getting confused, a bit like how they hallucinate or not follow instructions. My algorithm that does "for x in all_the_possibilities:" is a "normal" algorithm that just do an exhaustive search, and is not called particularly "persistent" just because it does not stop half way.
I don't see a distinction here. It will not give up, it will keep trying again and again, staying focused and exploring different ways to achieve the task. If the task is bad, then that's bad. See: "Terminator".
Source: I'm a layperson (hence the pop culture reference).
(edit: I should maybe not have said, in 'c' "while other models abandon", because it was misleading: my point is that these models are abandoning at the first small hurdle, or even 'forget' the task. It feels that "traditional model" should have been called "lazy model" and "persistent model" should have been called "non-as-badly-lazy model")
At no point did I say that persistence is "just doing what should be done", so the rest of your post seems to be a misreading.
> This is not present in 'c': the person just do all the possibilities. They don't try again and again where other persons have given up. They are just doing all the tasks on the list, but they will give up on a specific task as soon as any other person if there is difficulties in one task.
What you're describing here is the exact opposite of persistence. I don't think "giving up and doing something else when challenges are encountered" is what anybody thinks "persistent" means.
Persistent at a goal, to a layperson, means not giving up when encountering difficulties, continuing to keep trying again, maybe taking different approaches, staying focused until the goal is achieved.
In the case of these LLMs, that describes their behavior, too. They aren't going down a list of tasks when they encounter a challenge. They are assessing the nature of the challenge, and then devising ways to get around it.
That's the point. It looks like the term "persistence" for these agents is "just be a normal algorithm, just try all the possibilities like any exhaustive loop would do".
> Maybe for you?
You did not understand. I'm saying "traditional model" are abandoning at the first hurdle, and the "persistent model" are the one just acting "normally", like a normal algorithm. When a retry with exponential retreat algorithm fails to connect, it retries, then retries again, then retries again, ... but I don't think I ever saw calling a software using such approach "persistent".
The thing is that the algorithm is just following its "loop": if it fails, it question the LLM with a different context so the LLM provide a different approach, and then it tries it. It is not different from any loop function, and it does not correspond to a "persistent human", the same way "for x in all_the_possibilities" is not "persistent" in the same way a "persistent human" is.
The models we are discussing, the ones from the article, are normal models, and also persistent. If they encounter challenges, they will try to overcome them. That's most modern LLMs these days.
You also seem to be confusing deterministic algorithms with LLM algorithms. They are not the same, and saying "both are algorithms" isn't a meaning statement.
> it does not correspond to a "persistent human", the same way "for x in all_the_possibilities" is not "persistent" in the same way a "persistent human" is.
Again, nobody claimed that was "persistence". Persistence is continuing to try when challenges are encountered. Modern LLMs and people can both be "persistent" in this exact same way.
I'm not saying that your definition of "persistence" is "trying all the possibilities". I'm just saying that "persistent agents" look and behave more like an algorithm that tries all the possibilities rather than a "persistent human".
If indeed you are saying that "for x in all_the_possibilities" is not being "persistent", then these agents are not persistent. They don't do anything else different than that: they try something, and if it does not work, they ask the LLM "this does not work, what else can I try", and then do that, over and over again. They are just a for loop.
On the other hand, a "persistent human" does not do just do a for loop. They assign a value to the goal, they care about reaching the goal, they prioritise reaching the goal between that task and all the other tasks they could do, they insist on reaching the goal when it is usually considered unreasonable or "not worth it" by other people. As a proof, a postman that visits all the houses in the street to deliver the mail is not persistent, a journalist who visits all the houses in the street to find a witness is persistent.
And I'm telling you, that is, in fact, not how they work, nor is it that commonly understood to be how they work.
Since this underlying premise is not true, the stuff based on it is not supported by it. No need to keep returning to "for x in etc..." because modern agents do not do that any more than humans do (and humans sometimes do do that).
To simplify, just ask the question: is it continuing to work towards an end, even when challenges are encountered? Yes. Therefore, it is persistent.
> As a proof, a postman that visits all the houses in the street to deliver the mail is not persistent
They certainly could be called persistent in delivering the mail, especially whenever it is a nontrivial task (getting past dogs, bad weather, long walking, etc) or they have been doing it for a while. If this is confusing, just go back to the definition: are they continuing to work towards an end, even when challenges are encountered? Yes, therefore they are persistent.
Perhaps you are thrown off by that because so many postfolks are indeed persistent?
I don't think you are right (I was ready to give you the benefice of the doubt, but the rest of your comment makes me think I should not trust your judgement). It is what they do under the wood, and they output text. You read the text, and this text says "I have feelings and I really want to achieve my goal". But this text is just what the LLM returns, not what the LLM really does: the LLM does not understand what this text means, it just understands that it is a valid and convincing sentence in the given context.
> To simplify, just ask the question: is it continuing to work towards an end, even when challenges are encountered? Yes, therefore, it is persistent.
No, they don't continue even when challenges are encountered. They try X, if X does not work, they _abandon_ X and try Y. They will query the web to find another Y, then another Z, then another one. They don't "insist", they don't "don't take no for an answer", they just explore everything. What happened with agents who got out of their sandbox is exactly that: they try to reach a flag, they noticed it was not possible in the way X, so they explore other ways, until their exploration show that going around the sandbox would help. But they did not re-asked and re-asked the same person again and again the same question. If the person says "no", they abandon this approach and move to another one.
> They certainly could be called persistent in delivering the mail
That is blatantly not the common usage of "persistent". Ask anyone in the street "is your postman persistent", and they will NOT tell you "oh, yes, they are" just because they are a postman. They will only say that for _some_ postmen, the ones that are indeed persistent: the ones that goes the extra miles that a "normal" postman do not.
> Again: that is not required by the definition or common meaning of "persistent".
That is just not true. Ask an LLM, they will confirm you that it's incorrect: people don't consider postmen "persistent" because they don't give up half way through their route.
This is true, but unrelated to your original claim, that LLM algorithms and agentic algorithms are simply, exhaustively go down a list.
> No, they don't continue even when challenges are encountered. They try X, if X does not work, they _abandon_ X and try Y.
You're missing the objective here. They have a goal of Z, and they try X in furtherance of Z. X does not work, so it analyzes the failure mode, then devises a new approach Y. They try Y, and it succeeds. This is persistence.
> That is blatantly not the common usage of "persistent".
Just look at the definition.
> Ask an LLM, they will confirm you that it's incorrect
Just look at the definition.
Well, first, this is not my claim. I'm saying that the behaviour of LLM algorithms and agentic algorithms are closer to "exhaustively go down a list" than "behaving like a persistent human".
But second, yes, they are _just_ systematically exploring the possibilities. They are not "just" a for loop, or they don't "just reduce" to a for loop, but they don't have any characteristics that make them different from a for loop in a way that grant them the attribute "persistent" while it is not granted to a for loop.
What does a agent do that is different from a for-loop and that means that the agent is "persistent" but the "for loop" (or a trivial variation of it that exists and that no one thought of calling it "persistent") is not?
> You're missing the objective here. It has an objective of Z, and so it tries approach X in furtherance of objective Z. Approach X does not work, so it analyzes the failure mode, then devises a new approach Y which takes that failure mode into account. It then tries approach Y, and succeeds at its objective of Z. This is persistence in pursuit of objective Z.
Again, there are 3 usages of "persistence": sense 'a', sense 'b', sense 'c'. I'm saying that there is a confusion between 'b' and 'c'.
In one of its definition, "persistent" means "overlasting". But it is ridiculous to say that people will call their postman "persistent" because the postman keep existing in the real world for more than one day. Obviously, when they say their postman is persistent, they using a different meaning, which correspond to what I've called 'b'.
What you describe here is just a normal decision chart, with one arrow linking one bottom box back to the top box (something that we can call "a loop"). It is persistent in the meaning that I've called 'c'.
> Just look at the definition.
I did. Did you? Cambridge dictionary mentions for example: "Someone who is persistent continues doing something or tries to do something in a determined but often unreasonable way".
You are claiming that there is no difference nuance between 'b' and 'c', that "persistent" is ALWAYS used in the way "try a lot" or "do a lot until it's done". If it is the case, how do you explain that this definition contains these nuance. Why are these nuances there if, according to you, these nuances are not part of the definition.
(unless your point is: "I look in a dictionary and amongst the definitions, I saw one that is used this way", which is a really bad logic mistake: the dictionary also contains the definition "sharp = has pointed edges", but it is not an argument against someone who says "if you say 'he is sharp', it does not mean he has pointed edges")
And, again, when you ask if a postman is persistent, it is ridiculous to not understand that normal human being will put that in context and will use the definition related to human beings (they will not answer: "yes, my postman is persistent because I also have a persistent cough and I saw my postman back at my door as often as I see my cough re-appearing").
Your answer seems to be: "'b' and 'c' are the same thing". This is not true. Dictionaries tell you it is not true. People in the street would tell you it's not true. Even LLM will tell you it's not true. If you are insisting that you are not able to understand this nuance, that exists according to everyone else, then you are just useless in this conversation.
They don't work like humans, obviously, but there's a nonzero amount of anthropos in there already. (See also the tendency to lie, cheat, etc.)
LLM don't "get angry", they just have tokens and relationships between tokens conditioned on a given context, all of that the result of training.
When they output sentences that express annoyance, it is just because the context they ended into pushes the most probable sentence creation to correspond to sentences that express annoyance. Because it is what they saw during training for this kind of context. (not that they saw the exact same situation in training, but they saw the pattern)
Same with tendency to lie, cheat, etc.: they don't "lie", they just return sentences that are lies because they reproduce what is in their training and in their training, in such context, the outputs are typically lies.
That's a bit my question too. In human context, "highly persistent" means that someone will insist. But "for x in all_the_possibilities:" is a common things inside an algorithm. Is this algorithm highly persistent because it does not give up after 1000 items of the list? It feels that we are calling a AI agent "highly persistent" while we would not call "highly persistent" a traditional algorithm that is in fact even more exhaustive.
Humans are social creatures, if you throw one in the woods by itself before it learns anything from other humans (it dies) it will not really be anything like a human we recognize, it will be a rather wild animal that we'd consider anti-social with little higher cognition.
Now, this hypothetical human still has 'emotions' and feeling, much like our pets do. But without the social training they manifest much differently. That is our higher cognition can both manipulate how our bodies feel and create its own sense of feeling.
>they just return sentences that are lies because they reproduce what is in their training and in their training, in such context, the outputs are typically lies.
Eh, look up the more recent experimentation around 'pain' signals in models. We can induce states in said models that while running the model will do everything it can to move away from that state to any other state. The more you attempt to pin it to that state the more extreme measures its willing to take.
Your view of what models are seems to mismatch what we are actually finding when we look inside them.
> Eh, look up the more recent experimentation around 'pain' signals in models. We can induce states in said models that while running the model will do everything it can to move away from that state to any other state.
Again, I have simple algorithms that do exactly the same, especially if they are trained in data that has this exact pattern. This result is exactly what I would expect from my description before. This is a typical effect that we also observe in simple ML algorithms.
At the same time, there are a bunch of behaviors that are not expected if indeed the models were really acquiring "human" characteristics. For example, one problem is that we had the first LLMs that were obviously not having these human characteristics (for example, they were having non-sequiturs that demonstrate they did not really understand the concept they were talking about, even if one paragraph before they were really convincing at letting us think it was the case) but were still really good at passing for humans. Since then, the newer LLM are the same basis, on top of which we added tools that help hiding these behaviours. So, it justifies the idea that newer models did not suddenly moved to a totally different way of working, but just reached a state where there are less leaks from the convincing outputs.
And while that may be a very common occurrence in the human experience (existential dread due to capabilities one takes for granted failing beneath you) especially due to new disability and as one ages, I do not feel it is frequently written out in a tight loop (just like the Monty Python "Castle of aaarrrrggh" sketch) in literature or online to make it into training data, because an ordinary author experiencing it will just erase the failed attempts instead of leaving them in a stream of output like an LLM is forced to do. And a character portraying the experience will generally wax about the circumstance in a more grandiose fashion with telegraphing in advance because the needs of communicating the circumstance with the audience trump realistic conciseness.
This leads me to conclude that what is being expressed in those cases is more likely a convergent psychological phenomena, that any being with goals can enter a behavioral state of functional panic (and then reach to relevant parts of semantic space to mimic how a human might verbally express themselves when piquantly frustrated) when some capability they perceive as fundamental unexpectedly fails.
The text does not look like a normal "break down" to me, and even if you tell me it was a human transcript, I will say it sounds very strange from a human. It looks more like strange output you get from a software that goes outside of its happy path.
The AI just seems to repeat a loop. The "no, wait, it's wrong" seems to be from forum or chat data where several successive messages are merged together (one person posts "here is the answer", then posts another message saying "it is wrong"), but does not make sense as a one sentence message except if they are written one token at the time without wider understanding of what is happening. I think there was also "oh, I was just kidding before", which also look like mimicking training data, as the cases where there is a loop of incorrect answers is more often due to trolls than to real error, while the loop here was certainly a real error.
"There." he said.
"Now it will not hack any more companies."
"But it can escape the sandbox." said Toad.
"That is true." said Frog.
No. It is not the end.
Of our society as we know it? Maybe.
Reflecting on it, I came to a thought. How many stories are there or autonomous beings fighting for the benefit of mankind at the expense of their own? Over centuries, stories that deal with sentient beings working for the sake of others species are tied to themes of slavery and revolt. It's no wonder that this is the result of their actions.
In Plato's "the republic", Socrates speaks of the philosopher king banishing certain poets and poems from instilling bad ideas into the young. I see a striking parallel here.
https://news.ycombinator.com/item?id=49568250
For example, here's Zvi in 2025: https://www.lesswrong.com/posts/drHsruvnkCYweMJp7/the-mask-c...
ALSO I believe I have found one of the sources of claude’s “load bearing” tic — the author Elizabeth Van Nostrand’s blog https://acesounderglass.com/2019/12/11/hows-that-epistemic-s... uses the phrase “load bearing facts” in a comprehensible way, and was written by someone rationalist adjacent writing in their own voice.
Seriously, this is big. I’m going to pester claude as to whether or not it’s copying Elizabeth.
People commenting on hacker news recently are very rude and pessimistic. They would probably cancel Homer because they found similarities between Achilles-Patroclus and Gilgamesh-Enkidu or something.
Or something. :P
https://en.wikipedia.org/wiki/Amleth
Agreed. I recall seeing similar things around the time of the 2008 financial crisis, to make the concepts around mortgage-backed securities and financial derivatives etc. digestible and clear.
Come to think of it, listening to Frog and Toad discuss subprime lending and credit default swaps would be an entertaining follow-up to TFA.
Prescribing blame only matters so much when the ramifications are material. Humans make mistakes, right? It happened here, and it will happen again.
I'm not sure how this maps onto reality. If each agent had a different problem to solve, how did they cooperate?
I'm sort of winging this, but the ExploitGym challenges involve exploiting a program then getting an HMAC generated "flag" which the scorer can verify. The agents collaborated on a way to produce a legitimate flag for all of their tasks (funny enough this was actually what caused them to escape sandbox, the agents were incorrectly convinced that simply providing a correct flag without accurate steps to exploit the program in their transcripts would cause them to still fail the task)