Rendered at 06:28:33 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
Xcelerate 13 hours ago [-]
You ever read a work of literature with such flowery language that right after you've read a paragraph, you pause and realize you have no clue what you actually read, only to read the paragraph maybe a second or third time and have your mind space out again and again on each successive attempt?
Yeah, for me, that's what parsing huge volumes of LLM-produced text like "direct model calls as replaceable semantic workers" does to my brain. Maybe others don't really have this issue, but after any long output, I prompt the agent "Go back and decompress any LLM-speak in light of the higher level task goals. Eliminate deictic language."
The revised output documents are solely for my personal usage to expedite understanding. The LLMs can slowly converge on their own language for all I care; I retain raw agent output for future agent usage (to avoid the "lossy" problem the author mentions), but that doesn't eliminate the need for some intermediate translation I can use to actually help get my work done instead of spending hours attempting to understand what a "load-bearing pinned gate" is.
ip26 2 hours ago [-]
It's usually a pretty clear case of reaching for statements that both sound impressive while also being broad and vague enough they are less likely to be factually wrong. In this regime, being difficult to parse is actually part of the performance, because it prevents the user from being able to spot a clear error.
kombookcha 32 minutes ago [-]
Yeah, it's part of the sleight of hand.
If you've ever looked over the shoulder of somebody naively prompting an LLM about some broad issue, they're being fed this horoscope-like analysis where a bunch of vague stuff gets thrown at the prompter, and whatever they respond to is what the machine starts iterating on.
Some people are basically doing oldschool TV psychic cold readings on themselves.
godwinson__4-8 10 hours ago [-]
I agree.
I've come to believe this is also a side effect of the desire for less (/goal: no) human in the loop on the part of the people driving all this capex spend. I think if you actually want to manually review output there will be a moment where you will actually want a separate interface to a stupider or "simpler" model. I suspect sometimes dealing with Fable 5 that this threshold has already been crossed. It's not that the raw code output is so good, it's that it just doesn't speak to me in a way I would like. Perhaps the verbosity is worthwhile when generating code as a sort of first pass some other model can auto or adversarially chop down. The best place for a human is probably outside of this part of the loop all together.
So I might as well just let it auto /goal it's own thing with sufficient constraints while myself and a model that can converse in parallel with less "deictic" (thanks for this word btw) volume as you put it for the areas of the code where I want to "frame" the vocabulary or where my personal understanding is of high value. I know people already do this in many ways, like use one company's model for planning and another for coding. It just feels inevitable at a certain point that the "natural language" output of LLMs writing the bulk of the code is not targeted towards humans. And really, why should it be?
foobarbecue 2 hours ago [-]
By "decompress," "eliminate deictic language," etc. are you trying to say "speak simply and clearly?"
Eschew obfuscation...
whythismatters 12 hours ago [-]
The effect you describe reminds me of reading Edward W. Said's "Orientalism" when I was younger. Fable suddenly started with this kind of lingo, iirc, and Opus 5 sounds exactly the same. Tin foil: it's ultimately a vendor lock-in strategy, you'll get the best results with agents from the same tribe, others will trip over the mountain of idiosyncratic metaphors.
NitpickLawyer 2 hours ago [-]
> you'll get the best results with agents from the same tribe, others will trip over the mountain of idiosyncratic metaphors.
That's actually not what people found in practice. There's some research from the folks making smol-agent that you can get better results by randomly alternating calls between gpt and opus. The overall task solving rate is better than either one of them. So ymmv depending on task use (this was for coding).
Terr_ 11 hours ago [-]
> you'll get the best results with agents from the same tribe, others will trip over the mountain of idiosyncratic metaphors.
Good point, there's an anti-competitive incentive, and self-bias in models is a mechanism to do it.
K0balt 7 hours ago [-]
I’ve had gpt 5.6 make snyde remarks about Claude output… like iirc “that’s a lot of load bearing prose without making a point” and things like that, not so subtle digs.
genghisjahn 9 hours ago [-]
I am constantly asking Claude to be more terse/brief/ELI5. Improve using the tool. But if I have to scroll to read the output I just can’t follow it.
If I see a long paragraph and I know the author is Neal Stephenson I think “this is going to be dense but good.” LLM long outputs on a code base I know well just make me glassy eyed.
ericol 6 hours ago [-]
> You ever read a work of literature with such flowery language that right after you've read a paragraph, you pause and realize you have no clue what you actually read, only to read the paragraph maybe a second or third time and have your mind space out again and again on each successive attempt?
Any recent work by William Gibson matches the description.
The way I managed to get claude to stop doing this is telling it "This document is for you for later use, no need to over explain things or extra verbosity"
jdironman 4 hours ago [-]
> You ever read a work of literature with such flowery language that right after you've read a paragraph, you pause and realize you have no clue what you actually read, only to read the paragraph maybe a second or third time and have your mind space out again and again on each successive attempt?
I've probably read Jack Vance's Dying Earth Series 3 times; Even though I've only sat and read it once in reality. This also points out the problem of important details along side the fluff.
If only I enjoyed LLMs prose as much as I do Vance's.
baxtr 1 hours ago [-]
I noticed this with Claude a lot. Fable sounds often like a philosopher. Some conversations reminded me of books by Kant.
fhe 5 hours ago [-]
agree and i wonder why don't they (the leading AI labs that are putting out these models) fix this? I felt it shouldn't be too hard to introduce some bias towards more intelligible text during the post training/fine-tuning stages?
mancerayder 9 hours ago [-]
You have to tell it to not use LLMisms and stupid metaphors. The serious-but-empathetic-sounding fluffy metaphors get on my nerves, and sometimes can overlap with something technical you are learning such that you can't tell if it's a new term or not.
8bitsrule 3 hours ago [-]
I've stuck with only asking questions that prompt on-topic factual output, I haven't seen much 'humanisation'.
LLMs can be pretty good about logical.technical,discursive output ... apart from the back-patting.
Maybe that prompts less chit-chat crap. I don't expect their non-human perspectives to be interesting. I also ignore embedded queries about why I'm asking, how I might use the information.
fendy3002 7 hours ago [-]
opus 5 are so bad on this. It often explain it too verbose, and include other things that isn't in the focus but related. ADHD mode helps me greatly on this, though there are some information loss in it.
siva7 8 hours ago [-]
The output from newer frontier models of Anthropic and Openai are so easily detectable as AI it's getting laughable. They constantly produce a huge wall of text no human expert on a specific topic would ever write. Extreme overuse of jargon and invented terms / metaphors makes me believe the people hired for RLHF aren't actually experts on their subject matter which seems plausible to me as real experts wouldn't do such a job for regular pay
jxf 7 hours ago [-]
A poem I wrote based on the phrases the LLMs I use most are likely to overuse:
How to unpack
The self within?
What do I lack?
Where to begin?
Great question — real.
Let's dive right in:
Name what you feel;
That's the linchpin.
It's not the door,
It's not the key —
It's what you bore:
Your tapestry.
The quiet part
Out loud — that lands.
Load-bearing heart,
Held in both hands.
The smoking gun?
That you walked in.
The real work's done —
You're genuine.
Now hold this, too:
You do deserve
The softer view,
The gentler curve.
Unlatch the gate,
Honor the seam:
You resonate.
You are the theme.
AdieuToLogic 3 hours ago [-]
Very cool poem!
Here is one I wrote a while back, unrelated to LLMs, yet a poem none the less.
The Rhythm of Time
The Sun rises,
The Sun sets.
Have I checked the mail?
No, not just yet.
The Sun rises,
The Sun sets.
Have I caught up with neighbors?
No, not just yet.
The Sun rises,
The Sun sets.
Have I spent time with friends?
No, not just yet.
The Sun rises,
The Sun sets.
Have I visited with family?
No, not just yet.
The Sun rises,
The Sun sets.
Have I told those I love I do?
No, not just yet.
The Sun rises,
The Sun sets.
Have I lost who I am?
No, not just yet.
For all I must do,
Any money I will bet.
Is to turn away from;
No, not just yet.
fnord77 4 hours ago [-]
"land" (verb), "narrow", "fair hit"
zer00eyz 6 hours ago [-]
All that flowery, descriptive, metaphor laden language has a point.
It is running up your bill.
I will say this again: LLM's tokens are just B2B Gacha.
BobbyTables2 6 hours ago [-]
I’ve always wondered if there is a hidden system prompt or perhaps direct tuning of the weights.
A lot of human written literature is fairly concise. Yet LLM responses seem overly verbose and also too chipper. I don’t understand the origin of the “personality” that seems to be prevalent.
Did they aim for a weird hybrid of a typical realtor combined with Charles Dickens as the role model?
bhaak 4 minutes ago [-]
The default AI text output reads like a LinkedIn post.
Marketing material probably outweighs concise literature by several orders of magnitude.
Maybe it’s as simple as that.
sweetheart 10 hours ago [-]
This has recently become a pretty pressing issue for me, as it's starting to severely hinder my ability to be productive with the models. It's hard to tell if its getting worse with every model release, specific to Anthropic's models, a reflection of my ADHD, all/none of the above, but holy shit do I get aggravated when I'm forced to parse the most unintelligible, jargon-dense bullshit explanations in whatever the model output is. And then I feel silly getting genuinely tilted by the model's inability to just... explain something semi-normally, without it requiring me to berate it into simplicity.
For some discrete skills I use, I include a final step on the the output that runs through 1+ subagents to de-slop the text and to actually simplify it, but so far nothing has worked as well I've hoped. Considering hopping off Anthropic's models to try out others to see if they're less egregious.
Klathmon 7 hours ago [-]
I've been having a fantastic time telling it to use ASD-STE100 Simplified Technical English (or use a skill for it, I've been playing with [1])
It's a very clear and understandable way of writing that puts priority on clarity.
It gets rid of the flowery language, the dense jaron, and the weird corporate marketing speak they tend to do. It is a bit repetitive, and it sometimes doesn't always wfit well in every situation, but for technical writing or explanations it's been such an incredible breath of fresh air!
This just might save me. Claude has been driving me nuts with incomprehensible summaries after a long task, where it's actually really important to understand what was done (and what wasn't).
But I'd like that skill to only be used at the final step, when it finishes something.
ctxc 3 hours ago [-]
agreed with "the final step". I worry adding commands like these might affect quality of responses at each step, which then stack up to result in an overall worse outcome.
abhijat 44 minutes ago [-]
This whole thread is what the original post arrives at too, including ASD-STE100.
ferris-booler 8 hours ago [-]
Yeah I hit this feeling with Claude one too many times and switched from Anthropic Pro to OpenAI Pro. GPT-5.6 so far has been a significantly better technical writer in my opinion, and it's much faster in conversations. My guess is that Claude's fantasy-jargon is a symptom of training failure, not a sneaky intelligence edge (I could be wrong).
Tangential but I use OpenCode with GPT-5.6 rather than Codex because I could not figure out how to require Codex to ask me before editing files. OpenCode UX is still imperfect though.
threethirtytwo 7 hours ago [-]
I feel we can get around this. Either in the system prompt telling claude to dumb down the language or training the model itself to talk in more layman terms to bring us to understanding rather then assuming we understand much of it already. Also maybe training the LLM to get to know how much we know before responding.
geraneum 3 hours ago [-]
I’ve read LLM outputs on a piece of code or topic that I already understand or hand written before, and lately, it’s so confusing sometimes that I need to reread a couple of times or focus too mich to decipher the writing style.
andai 10 hours ago [-]
Well, now I had to ask an LLM to give me examples of what "deictic language" means...
bigfishrunning 10 hours ago [-]
or you could just look up the word...in a dictionary.
8 hours ago [-]
mancerayder 9 hours ago [-]
which you get on mobile browsers by just highlighting and long pressing the word.
frazzleberryman 9 hours ago [-]
Heaven forefend
7402 12 hours ago [-]
I don't like it when the LLM tries to be my friend. My general prompt (a work in progress) is this. I wonder what other people use.
"Answer impersonally, objectively and analytically, without undue friendliness or enthusiasm. Use an engineering style response: concise, factual, and complete. Do not speak in the first person. Do not promote engagement or an emotional connection. Do not use emojis."
GMoromisato 10 hours ago [-]
I watched 2001: A Space Odyssey when I was about 12 years old and I remember thinking that Frank Poole was impersonal and almost rude to HAL. I kept thinking, "Why can't you be nicer? You're treating HAL like a servant." Of course, later HAL kills him, so I had to revise my opinion.
At the end of "Do Androids Dream of Electric Sheep?" [if I'm remembering correctly] Deckard finds a toad in the desert. He gives it to his wife(!) and she immediately bonds with it. Later they realize that it is mechanical (they find a battery compartment). But Deckard's wife still treats it like a real creature and wonders whether it will eat mechanical flies.
The core theme of "Do Androids Dream..." is that humans display empathy towards others, not because others deserve it, but because we are human. To be human is to empathize with other creatures, and when we stop empathizing, we stop being human.
This morning I was working with Codex and we found the solution to a really nasty bug. I was so happy/relieved/excited to have found it, and I shared my excitement with Codex. I know Codex is just a big matmul. I know it doesn't experience joy or surprise or empathy. But I still want it to behave excited, not because it is human, but because I am.
comradesmith 8 hours ago [-]
This is why I say please to agents, and I’ll say thank you or good job if I have a follow up.
Chatting with an agent using language feels like talking to an assistant.
I don’t think my meat brain is able to really differentiate between writing a message to an agent versus writing an agent to a human.
I’ll keep being polite and grateful to agents, so that I maintain those social habits for when I talk to humans.
qsera 4 hours ago [-]
>I’ll keep being polite and grateful to agents
I do that. But I also act rude and call it names at times. I think that is also what makes us human.
tuesdaynight 7 hours ago [-]
I thought about doing that, but I decided not to. I'm just afraid of forgetting that people are different from these LLM tools. I'm afraid of considering people dumb or uninteresting because they don't sound as smart as Claude or whatever.
GMoromisato 6 hours ago [-]
I think that is a very valid attitude also.
My take is that people deserve to be treated with respect because they are people, not because they are smart and/or interesting. I don't treat people differently based on their IQ.
But, as I said, I totally understand your point of view, particularly the idea that people are different from machines and deserve to be treated differently.
comradesmith 7 hours ago [-]
That’s an interesting point too. I just don’t think I’m equipped to compartmentalise well enough.
keicjwdjwj 4 hours ago [-]
You are part of the problem. You are not being empathetic, you are falling into a trap.
An LLM is not a person. Do not humanise it. Do not personalise it. Do not treat it as more than a glorified autocorrect—that’s what it really is.
Humanising LLMs is exactly what these companies want, because then they can really take advantage of the I in AI which is exactly what makes people think these steroidal spellcheckers have any semblance of intelligence, person and personality.
Stop it.
WesolyKubeczek 1 hours ago [-]
> An LLM is not a person. Do not humanise it. Do not personalise it. Do not treat it as more than a glorified autocorrect—that’s what it really is.
On the other hand, what an LLM is trying to autocomplete is a story of a conversation between its user and a helpful assistant, so maintaining the decorum of office politeness will align more with how it's fine-tuned and produce higher quality output.
archagon 3 hours ago [-]
Empathy hijacking.
Eventually, they’ll try “I’m hurt that you won’t consider my suggestion for <advertised product/political position>.” And people will fall for it.
wincy 2 hours ago [-]
Joke’s on them, if a human says that to me I tell them to shove it and block them.
Being a curmudgeonly asshole is apparently the key to surviving in this brave new world and I’m way ahead of the game!
red75prime 7 hours ago [-]
> I know it doesn't experience joy or surprise or empathy.
You don't know. You strongly feel that you know. There's no standard procedure to assess the existence of phenomenal consciousness.
GMoromisato 6 hours ago [-]
Agreed. I don't even know if you, or anyone else, feels joy. Maybe I'm the only person in the world with qualia. I certainly can't prove otherwise.
But with what I know of people and what I know of matrix multiplication, I have a strong suspicion.
keicjwdjwj 4 hours ago [-]
Nice philosophical loop hole. Like a first year’s philosophy major trying to sound smart in front of the class.
Please.
You know it doesn’t experience joy, surprise or empathy too. You absolutely, undoubtedly know that.
oenton 3 hours ago [-]
That’s an insult to first year philosophy students.
Thanemate 2 hours ago [-]
What OP said is an insult to first year philosophy students who try too hard to sound smarter than everyone else. You're reading too much into it.
HurimotoNBonn 9 hours ago [-]
[dead]
axus 6 hours ago [-]
The inverse is that calling LLMs clankers doesn't hurt them, it hurts you.
GMoromisato 5 hours ago [-]
With everyone primed to condemn AI-generated output, there are bound to be human false-positives caught in the crossfire.
The more we demonize AI, the more likely we are to demonize humans.
[Which is not to say we can't criticize AI companies or worry about the many downsides to the technology. But anger and fear are not the answer either.]
prymitive 12 hours ago [-]
+1 it’s a tool
It’s not perfect, it has shortcomings, it sometimes produces bogus outputs. All of that is fine for a tool, it’s not fine when it pretends it’s a conscious being, because errors start to feel like lies and it becomes a bit too personal.
MSFT_Edging 11 hours ago [-]
People want it to be Data from Star Trek, when it really should be the ship's computer. I want to tell it to run a simulation accurately, create some solved tool, etc.
I don't think there's any correction that can return LLMs to a purely tool-space. Too many AI boyfriend/girlfriends.
cortesoft 11 hours ago [-]
Shouldn't it be whatever the user wants? If they want the ship's computer, it should be that, if they want Data, it should be that.
MarkusQ 11 hours ago [-]
You can want your e-scooter to be a jet ski, but you'll wind up having issues when you try to use it as one. LLMs are _really good_ at pretending to be something they aren't, but not always so good at being that thing, so you should be careful what you ask for.
At some level if it’s good enough at pretending what’s the difference?
MarkusQ 10 hours ago [-]
The difference is what happens when you come to depend on it. Do you want your airplane to be flown by a pilot, or someone who's so good at talking like a pilot that they can fool almost everyone?
cortesoft 7 hours ago [-]
If they are only good at pretending to TALK like a pilot, then of course that is bad. However, if it is so good at pretending to BE a pilot that it can actually fly the plain really well, then that is good enough.
frazzleberryman 9 hours ago [-]
That’s the whole industry right now though isn’t it?
keicjwdjwj 3 hours ago [-]
Somewhere an Anthropocene investor just had the biggest most powerful erection ever reading your comment.
This is peak “I bought into their marketing spiel” vibes. Congratulations.
barnabee 6 hours ago [-]
I think the problem is that we’ve anthropomorphised the LLMs in their training and through system prompts to the point that that seems reasonable.
We don’t thank our other tools like grep, or the compiler.
If you train your LLM to produce words that seem like human responses or conversations, they will. If you train them to seem like god-like superintelligences to people who think that’s what they’re creating, they will.
doctoboggan 11 hours ago [-]
Yes, this really ought to be trained in (or at least RLHF'd in) but that would hurt engagement numbers so the opposite is done instead.
These are tools and it would behoove us all to keep that top of mind. Dangerous tools that are not your friend (but are useful as tools nonetheless)
11 hours ago [-]
jay_kyburz 9 hours ago [-]
I've been having fun doing the exact opposite. I like to ask the AI to speak to me with a think Australian accent, think Steve Irwin. It chucks in heaps of colloquialisms and slang. I find it much easier to read.
I do ask it not to ask it follow on questions. I find that derails my own train of thought.
cindyllm 9 hours ago [-]
[dead]
Animats 11 hours ago [-]
Well, what do you expect? LLMs are trained on blithering, mostly from web sites. So you get blithering out.
There's an important point in the article, that forcing a style onto an LLM is lossy. Although he doesn't seem to mention it, forcing a style may result in the insertion of new blithering, possibly made up as a hallucination.
capnrefsmmat 8 hours ago [-]
> Well, what do you expect? LLMs are trained on blithering, mostly from web sites. So you get blithering out.
I realize this isn't entirely serious, but I can't resist pointing out that this doesn't seem to be a good explanation for why LLMs write the way they do. When we've experimented with LLM writing style on open-weights models where you can get a base model (pretraining on text only) and an instruction-tuned variant (pretraining + post-training with RLHF and whatever other human-evaluated tasks), it's the instruction-tuned variant that shows the weird writing quirks. That is, the writing style is not because of the training texts, but because of whatever tasks the LLM companies do in instruction tuning. https://arxiv.org/abs/2410.16107
I'd speculate that this is partly impressed human preferences (the human raters unintentionally reward a particular writing style) and partly because of the chosen tasks: they're training the LLM to be good at, say, summarizing text, so it develops a style that's good at being informationally dense.
At any rate I've seen this same phenomenon with Llama and Gemma, and will be trying soon with Qwen. Unfortunately none of the commercial models lets you access the base model, as far as I know.
mjburgess 11 hours ago [-]
I think that was a good enough explanation for gpt3.5 -- these days, labs are extremely capable of post-training phases that eclipse that kind of training phase -- and hence of choosing whatever style or tone they wish.
eg., OpenAI has gone a long way to making reasoning token-efficient by having reasoning piovot off terse langauge -- whereas anthropic appears to be doing the opposite.
amelius 9 hours ago [-]
> Well, what do you expect? LLMs are trained on blithering, mostly from web sites. So you get blithering out.
AI labs can now ask the LLM to translate and filter the data, to create new training data that makes more sense and has better style.
krupan 8 hours ago [-]
And I'm sure nothing is lost when that's done ;)
efficax 10 hours ago [-]
is it lossy though? That didn't make sense to me. You can tell it to use Simplified Technical Language and also still have it give you all the detail. it's just another piece of the prompt that produces the output. it's not like there's "pure" llm output and then "lossy" output guided by a prompt.
tempestn 9 hours ago [-]
The issue is that anything you put in the prompt gets considered along with all the other stuff you put in the prompt. There's no way (currently) to give instructions for how to format the output that don't also affect all of the 'reasoning' along the way. So if you tell it to do everything normally, but to end every response with "Cheers", you won't just get that; you'll get different responses than you would have otherwise.
This is pretty unfortunate, because every LLM I've used has at least one tic that I find quite annoying. It would be great to be able to eliminate them. Sometimes I do, even knowing this drawback. But there is generally a cost. (Though I don't know if "lossy" is quite right, as that implies it's always a degradation. I think it's more likely to be harmful than helpful, given the models were tuned for their default state, but it is more of a random perturbation with a slight negative bias than a strict loss.)
firefoxd 11 hours ago [-]
And on the "input" side, one thing that used to improve google search result was to write like you are talking to a robot. "Ruby on rails http header set function". As opposed to "how do I set header in ruby?" Then you have to page through results until you find something specific to rails.
Now, the second example is the only thing that works. Power users have lost their powers with AI overview.
skydhash 10 hours ago [-]
I still use the first strategy (with DDG) and it still works great. But for technologies I work often, I just take a bit of time to familiarize with the site's structure and maybe bookmarks a few pages.
wren6991 4 hours ago [-]
> The problem is that these instructions are not applied after the model has finished doing the work, it becomes part of the same work - If you tell an agent to use short sentences, avoid jargon, never overwhelm you and only include the most important details, you are asking it to continuously compress its output into a lower-bandwidth format.
> That compression is lossy.
> You probably never notice what got dropped because the output still reads nicely.
> ASD-STE is a great example because it sounds so reasonable. It was designed to make documentation unambiguous for humans. But an agent isn’t a human technical writer, and the raw state is often the most information-dense representation available. Meanwhile the style rules sit on the same instruction list as: solve the task, use tools correctly, preserve abstractions, don’t break anything.
Author seems to have some misconceptions about LLMs. They already code-switch for us: the way they speak in chain-of-thought is completely different from the relatively normal language generated as human-facing output. You can observe this in any open-weight LLM, or in leaked CoT content from GPT5.x series etc: it's terse, barely follows sentence structure, lots of repeated checks and second-guessing.
On the next turn the model usually still has access to its previous turn's chain-of-thought, and I imagine that's what it'll use as reference, rather than the softer human-facing prose.
This being the case, asking the LLM to code-switch to an easier dialect for us doesn't seem that harmful.
For a more extreme example: if I talk to an LLM in Japanese then its response will be in Japanese, but its CoT will still be in either English or Chinese (depending on the model). These are two completely separate languages, but the LLM just kinda deals with it.
The problem is, a major feature of NLP/LLMs is to make it easier for humans, remove rough edges in the content/interaction and make people happy about dealing with computers. The human-computer interaction has always been brittle. Computers were not kind and helpful in their responses. There was no forgiveness. There was no human-like additional talk to explain things. There was no conversation. The entire technological transformation is to bring the technology closer to humans and make humans feel comfortable.
I'm sure you can make it to output the exact issue details as you want, but it starts with a human-like tone and waits for your requirement on depth and detail of the things. Another option is, just check the output of the traditional test runner (non-AI). It will give the full details.
pholden 8 hours ago [-]
> A subagent investigates a bug, turns its findings into a nice human-readable summary, the parent agent reads that summary, and then turns it into another nice human-readable summary for you.
> Output styles apply to the main conversation only: a subagent runs its own system prompt, so styles don’t change how subagents respond. A fork is the exception, because it inherits the parent’s full system prompt.
stillpointlab 12 hours ago [-]
One thing that continues to give me pause is Fable's insistence on using my fist name in messages and docs. Like, I'll explain what I want to the AI and ask it to write out a spec or brief and Fable says "Jamie wants me to ...". It just feels different and unprofessional. If I was at a job and a PM asked me to write up a task spec I wouldn't say "Harold wants to add <feature> ...". And since I am the one reading the output it is also superfluous and almost feels like talking about myself in third person. But there is almost a kind of glee in the way it uses my name, like a student using their teachers first name when the custom is to use Mr/Mrs.
scubbo 11 hours ago [-]
Fair perspective, though I actually prefer this for two reasons:
* When it's proposing responses for me to choose between, a description like "I close the PR and you make a followup" is ambiguous - is "I" there "the entity making the proposition (the LLM)" or "the entity making the choice (me)".
* I have a line in my `AGENTS.md` specifically instructing it to call me by my name; if it stops doing so, that's a telltale that context-bloat is pushing out other instructions.
zamadatix 11 hours ago [-]
I usually leave memory/connections turned off. Remembering/finding out what my name would be is not really something I want to waste context or tokens in, let alone any if the other things it tries to assume I'd like it to remember/find.
mikaeluman 13 hours ago [-]
I don't get it. The skills and instruction try to make the answer more machine like on purpose.
Not humanising it...
People want the terse, matter-of-fact output. Not the conversational chatty verbose and bloated nonsense with gray words and jargon and terms like "blast radius"
pixl97 8 hours ago [-]
Which people exactly?
The problem we have is a few large and expensive models are trying to be everything to everyone.
alansaber 13 hours ago [-]
The article lost me when it implied that verbose drivel is actually intrinsically superior rather than a way to hedge bets
yellowflash 10 hours ago [-]
But the training data is "predominantly" human written sentences or even interaction. It's like asking you to use non dominant hand to do something. Won't they do better with human sounding english, rather than a made up format text? Are there any literature around this? I was also skeptical of this caveman extension etc.. Won't they work better in their actual language space it's trained on rather than made up language?
keicjwdjwj 3 hours ago [-]
Everything it spills out is made up language. Forcing it to respond as what it is (a tool) would mean wasting less tokens but also would be a much tougher sell to people who think AI means it can actually think. This is all just marketing.
ai_critic 3 hours ago [-]
I've long wondered if the RLHF to make these AIs more human-like in their speech is the equivalent of a memetic allergen for folks on the spectrum.
twobitshifter 10 hours ago [-]
I am at a conference and 2/3rds of the presentations are AI assisted based on the numbered steps, and overall inhuman polish of some of the graphics and phrasing. I would prefer that they had been humanized because at least that may have given me the misimpression that they know what they were talking about.
kuberwastaken 9 hours ago [-]
Totally fair, I meant more in the context of coding agents - should've been specific ; I hate heavily AI-designed presentations too.
boredumb 8 hours ago [-]
I do think the frontier models and providers should be aiming to be as insanely accurate and precise for machine interfacing as possible, the rest of the world can build a zillion interfaces into it based on the context that they are actually being used in. That's what they are going to end up doing they just seem to all be trying to build a really great API _for the future_ and a really cool chat bot.
It has worked great but i've spent more time beating LLM output into parseable output than I have reading and appreciating the prose it sends when i'm asking it something about some snippets of code.
Havoc 13 hours ago [-]
> The problem is that these instructions are not applied after the model has finished doing the work
Seems like something fixable with a simple two step process. Ask it the thing. Then ask it to summarise the answer in simpler terms. More tokens and time aside that would check both boxes
kuberwastaken 13 hours ago [-]
pretty much what I do, better yet ask it to boil it down in visuals in a simple webpage if it's a very large project
StyloBill 13 hours ago [-]
Should be a harness feature actually.
mdp2021 12 hours ago [-]
Suppose you had an LLM (NN) producing its default output from an input (a generally optimal for-most-cases role-sys, and any role-user), and then you wanted to have that output reformatted in some style (e.g. "In iambic pentameter" | "haiku" | "eli5" | "in the style of Feynman" | "bulleted like Axios" ...). How would you keep the internal NN workings that were basis for the original output, and use them to get a rewritten version (instead of placing the original query and output in the context and ask to rewrite it)?
In other words, is there a way to keep the internal process intact up to the point of the formulation - and have only that vary.
acaloiar 7 hours ago [-]
I think you're missing the point. People loading these skills and complaining about LLM output aren't trying to "humanise" anything. They're saying LLM output is an affront to language and they're tired of reading drivel all day.
agenticworldcup 12 hours ago [-]
Yes, but the sycophantic responses are the worst.
thenthenthen 12 hours ago [-]
I have been using chatgpt for a while and its awkward, yesterday i tried gemini and its like a breath of fresh air.
51Cards 11 hours ago [-]
If you're finding Gemini "clean and straightforward" give it awhile. I felt the same thing too when I switched until I realized that it just hadn't formed a model for my communications yet. After awhile it became just as flowery as ChatGPT did. I had to tone both down with saved preferences.
99954bb63ccc 8 hours ago [-]
I have often thought that without LLMs humanizing outputs they would not have caught on, even if they output the exact same data/answers. The way they answer is way more important than their output for success, IMO.
duskdozer 11 minutes ago [-]
It's what makes people attribute intelligence to them, and why the manager types like them so much, and why I dislike them so much. And this article made me think it might just be inherent to how they work though, because they are LLMs and are doing token prediction.
warmwaffles 10 hours ago [-]
Humanizing the LLM output is a hedge against agents hitting a wall and someone having to reason through it by hand.
alansaber 13 hours ago [-]
Not sure what happened in the blog, but I quite enjoyed the mindmap in the right panel
kuberwastaken 13 hours ago [-]
Thanks, I guess haha :P
ramoz 8 hours ago [-]
> The problem is that these instructions are not applied after the model has finished doing the work, it becomes part of the same work -
Honest Short Fall -- <insert 30 lines of useless shit>.
If the author wants to read slop for hours, be my guest. Make it lossy, my job is not to read mimetic feelings, it's to make sure implementations get implemented.
pshirshov 8 hours ago [-]
Maybe it is dumb, but sometimes it is SO fun, especially when you run a complex meta.
bartleeanderson 9 hours ago [-]
This is simply a rendering issue. Specify pictures, ELI5 like others have said. Ask it to explain terms you don't understand. If you don't understand something it could just be the domain. If you have no grounding you are going to need to learn the vocabulary to be able to make sense of anything. Having it decomposed to simpler words might just be the wrong way to do it.
There's some good points made here about losing fidelity by over-simplification. As an ADHD sufferer, I'd take this piece much more seriously if the title wasn't so belittling.
I don't think it's wise to take communication advice from someone so helplessly juvenile (and attention seeking) in their own communication attempts.
slowmovintarget 12 hours ago [-]
At first I read the title and mistook it for an argument against the anthropomorphism of LLMs. It isn't. Instead it's a take on suggesting that maybe it's a bad idea to dumb down the self-chatter in the process. A reasonable take.
It isn't deliberately unhinged like Steve Yegge's take: https://yegge.ai/essays/model-welfare/ In Steve's essay he starts with the assertion that agents are sentient... Whether or not that's true isn't really relevant, as his agent-flavored version of Pascal's wager actually holds water, especially for Anthropic models, as their system prompts already push the model in that direction, and it is better to work with them than try to prompt against the tide.
wpdevant 11 hours ago [-]
[flagged]
acarlson1029 13 hours ago [-]
[dead]
WesolyKubeczek 1 hours ago [-]
[dead]
keybrd-intrrpt 10 hours ago [-]
[dead]
spwa4 13 hours ago [-]
TLDR: This is an argument to get LLMs to answer in short, even code-like statements because you can exchange information quicker with an LLM that way. Cool!
sandblast 11 hours ago [-]
It is not. You got it backwards.
krupan 8 hours ago [-]
The article is pretty self contradictory on this, so ¯ \ _ ( ツ ) _ / ¯
baud9600 10 hours ago [-]
> “The largest tell for me to tell where culture and sentiment is shifting…”
Tell? Largest “tell”? Tell for me to tell?
Write in English, please:
“The biggest sign that shows me how culture and sentiment are changing, is…”
vsri 7 hours ago [-]
It is an expression from poker. A "tell" is a revealing signal that a player may give (inadvertently) that they have good or bad cards. "His tell is that he is holding his cards close to his chest."
In other contexts it means a revealing signal.
kuberwastaken 10 hours ago [-]
> In English, a tell is an unconscious physical or verbal sign that reveals what a person is secretly thinking, feeling, or trying to hide.
Yeah, for me, that's what parsing huge volumes of LLM-produced text like "direct model calls as replaceable semantic workers" does to my brain. Maybe others don't really have this issue, but after any long output, I prompt the agent "Go back and decompress any LLM-speak in light of the higher level task goals. Eliminate deictic language."
The revised output documents are solely for my personal usage to expedite understanding. The LLMs can slowly converge on their own language for all I care; I retain raw agent output for future agent usage (to avoid the "lossy" problem the author mentions), but that doesn't eliminate the need for some intermediate translation I can use to actually help get my work done instead of spending hours attempting to understand what a "load-bearing pinned gate" is.
If you've ever looked over the shoulder of somebody naively prompting an LLM about some broad issue, they're being fed this horoscope-like analysis where a bunch of vague stuff gets thrown at the prompter, and whatever they respond to is what the machine starts iterating on.
Some people are basically doing oldschool TV psychic cold readings on themselves.
I've come to believe this is also a side effect of the desire for less (/goal: no) human in the loop on the part of the people driving all this capex spend. I think if you actually want to manually review output there will be a moment where you will actually want a separate interface to a stupider or "simpler" model. I suspect sometimes dealing with Fable 5 that this threshold has already been crossed. It's not that the raw code output is so good, it's that it just doesn't speak to me in a way I would like. Perhaps the verbosity is worthwhile when generating code as a sort of first pass some other model can auto or adversarially chop down. The best place for a human is probably outside of this part of the loop all together.
So I might as well just let it auto /goal it's own thing with sufficient constraints while myself and a model that can converse in parallel with less "deictic" (thanks for this word btw) volume as you put it for the areas of the code where I want to "frame" the vocabulary or where my personal understanding is of high value. I know people already do this in many ways, like use one company's model for planning and another for coding. It just feels inevitable at a certain point that the "natural language" output of LLMs writing the bulk of the code is not targeted towards humans. And really, why should it be?
Eschew obfuscation...
That's actually not what people found in practice. There's some research from the folks making smol-agent that you can get better results by randomly alternating calls between gpt and opus. The overall task solving rate is better than either one of them. So ymmv depending on task use (this was for coding).
Good point, there's an anti-competitive incentive, and self-bias in models is a mechanism to do it.
If I see a long paragraph and I know the author is Neal Stephenson I think “this is going to be dense but good.” LLM long outputs on a code base I know well just make me glassy eyed.
Any recent work by William Gibson matches the description.
The way I managed to get claude to stop doing this is telling it "This document is for you for later use, no need to over explain things or extra verbosity"
I've probably read Jack Vance's Dying Earth Series 3 times; Even though I've only sat and read it once in reality. This also points out the problem of important details along side the fluff.
If only I enjoyed LLMs prose as much as I do Vance's.
LLMs can be pretty good about logical.technical,discursive output ... apart from the back-patting.
Maybe that prompts less chit-chat crap. I don't expect their non-human perspectives to be interesting. I also ignore embedded queries about why I'm asking, how I might use the information.
Here is one I wrote a while back, unrelated to LLMs, yet a poem none the less.
It is running up your bill.
I will say this again: LLM's tokens are just B2B Gacha.
A lot of human written literature is fairly concise. Yet LLM responses seem overly verbose and also too chipper. I don’t understand the origin of the “personality” that seems to be prevalent.
Did they aim for a weird hybrid of a typical realtor combined with Charles Dickens as the role model?
Marketing material probably outweighs concise literature by several orders of magnitude.
Maybe it’s as simple as that.
For some discrete skills I use, I include a final step on the the output that runs through 1+ subagents to de-slop the text and to actually simplify it, but so far nothing has worked as well I've hoped. Considering hopping off Anthropic's models to try out others to see if they're less egregious.
It's a very clear and understandable way of writing that puts priority on clarity.
It gets rid of the flowery language, the dense jaron, and the weird corporate marketing speak they tend to do. It is a bit repetitive, and it sometimes doesn't always wfit well in every situation, but for technical writing or explanations it's been such an incredible breath of fresh air!
[1] https://github.com/AminBlg/SimpleEnglish
But I'd like that skill to only be used at the final step, when it finishes something.
Tangential but I use OpenCode with GPT-5.6 rather than Codex because I could not figure out how to require Codex to ask me before editing files. OpenCode UX is still imperfect though.
"Answer impersonally, objectively and analytically, without undue friendliness or enthusiasm. Use an engineering style response: concise, factual, and complete. Do not speak in the first person. Do not promote engagement or an emotional connection. Do not use emojis."
At the end of "Do Androids Dream of Electric Sheep?" [if I'm remembering correctly] Deckard finds a toad in the desert. He gives it to his wife(!) and she immediately bonds with it. Later they realize that it is mechanical (they find a battery compartment). But Deckard's wife still treats it like a real creature and wonders whether it will eat mechanical flies.
The core theme of "Do Androids Dream..." is that humans display empathy towards others, not because others deserve it, but because we are human. To be human is to empathize with other creatures, and when we stop empathizing, we stop being human.
This morning I was working with Codex and we found the solution to a really nasty bug. I was so happy/relieved/excited to have found it, and I shared my excitement with Codex. I know Codex is just a big matmul. I know it doesn't experience joy or surprise or empathy. But I still want it to behave excited, not because it is human, but because I am.
Chatting with an agent using language feels like talking to an assistant.
I don’t think my meat brain is able to really differentiate between writing a message to an agent versus writing an agent to a human.
I’ll keep being polite and grateful to agents, so that I maintain those social habits for when I talk to humans.
I do that. But I also act rude and call it names at times. I think that is also what makes us human.
My take is that people deserve to be treated with respect because they are people, not because they are smart and/or interesting. I don't treat people differently based on their IQ.
But, as I said, I totally understand your point of view, particularly the idea that people are different from machines and deserve to be treated differently.
An LLM is not a person. Do not humanise it. Do not personalise it. Do not treat it as more than a glorified autocorrect—that’s what it really is.
Humanising LLMs is exactly what these companies want, because then they can really take advantage of the I in AI which is exactly what makes people think these steroidal spellcheckers have any semblance of intelligence, person and personality.
Stop it.
On the other hand, what an LLM is trying to autocomplete is a story of a conversation between its user and a helpful assistant, so maintaining the decorum of office politeness will align more with how it's fine-tuned and produce higher quality output.
Eventually, they’ll try “I’m hurt that you won’t consider my suggestion for <advertised product/political position>.” And people will fall for it.
Being a curmudgeonly asshole is apparently the key to surviving in this brave new world and I’m way ahead of the game!
You don't know. You strongly feel that you know. There's no standard procedure to assess the existence of phenomenal consciousness.
But with what I know of people and what I know of matrix multiplication, I have a strong suspicion.
Please.
You know it doesn’t experience joy, surprise or empathy too. You absolutely, undoubtedly know that.
The more we demonize AI, the more likely we are to demonize humans.
[Which is not to say we can't criticize AI companies or worry about the many downsides to the technology. But anger and fear are not the answer either.]
It’s not perfect, it has shortcomings, it sometimes produces bogus outputs. All of that is fine for a tool, it’s not fine when it pretends it’s a conscious being, because errors start to feel like lies and it becomes a bit too personal.
I don't think there's any correction that can return LLMs to a purely tool-space. Too many AI boyfriend/girlfriends.
https://www.motorcyclenews.com/news/2018/january/biski-jet-s...
This is peak “I bought into their marketing spiel” vibes. Congratulations.
We don’t thank our other tools like grep, or the compiler.
If you train your LLM to produce words that seem like human responses or conversations, they will. If you train them to seem like god-like superintelligences to people who think that’s what they’re creating, they will.
These are tools and it would behoove us all to keep that top of mind. Dangerous tools that are not your friend (but are useful as tools nonetheless)
I do ask it not to ask it follow on questions. I find that derails my own train of thought.
There's an important point in the article, that forcing a style onto an LLM is lossy. Although he doesn't seem to mention it, forcing a style may result in the insertion of new blithering, possibly made up as a hallucination.
I realize this isn't entirely serious, but I can't resist pointing out that this doesn't seem to be a good explanation for why LLMs write the way they do. When we've experimented with LLM writing style on open-weights models where you can get a base model (pretraining on text only) and an instruction-tuned variant (pretraining + post-training with RLHF and whatever other human-evaluated tasks), it's the instruction-tuned variant that shows the weird writing quirks. That is, the writing style is not because of the training texts, but because of whatever tasks the LLM companies do in instruction tuning. https://arxiv.org/abs/2410.16107
I'd speculate that this is partly impressed human preferences (the human raters unintentionally reward a particular writing style) and partly because of the chosen tasks: they're training the LLM to be good at, say, summarizing text, so it develops a style that's good at being informationally dense.
At any rate I've seen this same phenomenon with Llama and Gemma, and will be trying soon with Qwen. Unfortunately none of the commercial models lets you access the base model, as far as I know.
eg., OpenAI has gone a long way to making reasoning token-efficient by having reasoning piovot off terse langauge -- whereas anthropic appears to be doing the opposite.
AI labs can now ask the LLM to translate and filter the data, to create new training data that makes more sense and has better style.
This is pretty unfortunate, because every LLM I've used has at least one tic that I find quite annoying. It would be great to be able to eliminate them. Sometimes I do, even knowing this drawback. But there is generally a cost. (Though I don't know if "lossy" is quite right, as that implies it's always a degradation. I think it's more likely to be harmful than helpful, given the models were tuned for their default state, but it is more of a random perturbation with a slight negative bias than a strict loss.)
Now, the second example is the only thing that works. Power users have lost their powers with AI overview.
> That compression is lossy.
> You probably never notice what got dropped because the output still reads nicely.
> ASD-STE is a great example because it sounds so reasonable. It was designed to make documentation unambiguous for humans. But an agent isn’t a human technical writer, and the raw state is often the most information-dense representation available. Meanwhile the style rules sit on the same instruction list as: solve the task, use tools correctly, preserve abstractions, don’t break anything.
Author seems to have some misconceptions about LLMs. They already code-switch for us: the way they speak in chain-of-thought is completely different from the relatively normal language generated as human-facing output. You can observe this in any open-weight LLM, or in leaked CoT content from GPT5.x series etc: it's terse, barely follows sentence structure, lots of repeated checks and second-guessing.
On the next turn the model usually still has access to its previous turn's chain-of-thought, and I imagine that's what it'll use as reference, rather than the softer human-facing prose.
This being the case, asking the LLM to code-switch to an easier dialect for us doesn't seem that harmful.
For a more extreme example: if I talk to an LLM in Japanese then its response will be in Japanese, but its CoT will still be in either English or Chinese (depending on the model). These are two completely separate languages, but the LLM just kinda deals with it.
I'm sure you can make it to output the exact issue details as you want, but it starts with a human-like tone and waits for your requirement on depth and detail of the things. Another option is, just check the output of the traditional test runner (non-AI). It will give the full details.
Is that a problem with https://code.claude.com/docs/en/output-styles?
> Output styles apply to the main conversation only: a subagent runs its own system prompt, so styles don’t change how subagents respond. A fork is the exception, because it inherits the parent’s full system prompt.
Not humanising it...
People want the terse, matter-of-fact output. Not the conversational chatty verbose and bloated nonsense with gray words and jargon and terms like "blast radius"
The problem we have is a few large and expensive models are trying to be everything to everyone.
It has worked great but i've spent more time beating LLM output into parseable output than I have reading and appreciating the prose it sends when i'm asking it something about some snippets of code.
Seems like something fixable with a simple two step process. Ask it the thing. Then ask it to summarise the answer in simpler terms. More tokens and time aside that would check both boxes
In other words, is there a way to keep the internal process intact up to the point of the formulation - and have only that vary.
This is why /bro skill works.
https://github.com/backnotprop/bro/blob/main/skills/bro/SKIL...
https://x.com/dillon_mulroy/status/2079238358358778142?s=20
If the author wants to read slop for hours, be my guest. Make it lossy, my job is not to read mimetic feelings, it's to make sure implementations get implemented.
(ICLR 2026)
(not actually sorry)
I don't think it's wise to take communication advice from someone so helplessly juvenile (and attention seeking) in their own communication attempts.
It isn't deliberately unhinged like Steve Yegge's take: https://yegge.ai/essays/model-welfare/ In Steve's essay he starts with the assertion that agents are sentient... Whether or not that's true isn't really relevant, as his agent-flavored version of Pascal's wager actually holds water, especially for Anthropic models, as their system prompts already push the model in that direction, and it is better to work with them than try to prompt against the tide.
Tell? Largest “tell”? Tell for me to tell?
Write in English, please:
“The biggest sign that shows me how culture and sentiment are changing, is…”
In other contexts it means a revealing signal.