Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Exhaustion of Talking to a Tool

ohadravid.github.io

Post reply on HN
Print view View original

51–60 of 85 posts

Re: The Exhaustion of Talking to a Tool

#51

Post by enraged_camel »

mym1990 wrote: post #10
enraged_camel wrote: post #6

>> Is it worth the social brainwork? IDK, for some tasks maybe - there are things a single person can do now that would have been impossible a year ago. But for all tasks? And wouldn’t that social brainwork do more good if it was directed at the real people you are working with? It's the opposite. Many people find it exhausting to interact with humans, and do so only because they are required to. Humans often don't u…

Fooling yourself into thinking the LLMs are "understanding" what you are saying or asking is a trap. The output you get may be useful, but it is not due to any sort of understanding. The elements of human work you mentioned are why it can be both rewarding and painful to interact with humans, but at the end of the day it is important to keep trying to do that work/keep trying to interact with each other. I don't know…

>> Fooling yourself into thinking the LLMs are "understanding" what you are saying or asking is a trap. The output you get may be useful, but it is not due to any sort of understanding.

I hear this a lot but I think it's a matter of semantics and ultimately not very useful. I don't care whether the LLM understands me the way a human would. I use the LLM to get useful output. I want it to do something and it does that thing.

Re: The Exhaustion of Talking to a Tool

#52

Post by Joel_Mckay »

mewpmewp2 wrote: post #38
Joel_Mckay wrote: post #36

Earlier quoted context omitted.

A web search doesn't use isomorphic plagiarism of naive users work to sell to other users. LLM are actually good at context search, but are mostly not being used as intended. The LLM hype bubble has to end sooner or later. =3

How does that answer the question or relate to variable reward and addiction mechanisms? This just seems another unrelated negative sentiment argument towards LLMs?

Not really, people are simply fooled by pernicious sycophantic idealism, and their own cognitive biases.

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

LLM are great for context search, but were never real "AI" in any sense except in media/PR sensationalism. Scientific hubris and unethical use of LLM is interesting as exposes the primitive impulsive nature of many. Best of luck =3

Re: The Exhaustion of Talking to a Tool

#53

Post by Joel_Mckay »

mewpmewp2 wrote: post #50
Joel_Mckay wrote: post #44

Earlier quoted context omitted.

Happiness is derived from a balance of a meaningful life, and pleasant experiences we may choose. Also, one may have no pain or concerns, and still existentially despair over a meaningless life built on intelligence campaigns exploiting millions of people. Only psychopaths find short lived joy in harming others, and only make up around 1% of general populations. Have a wonderful day =3

I think happiness is very complicated. Would you think there's a perfect formula to follow that will yield you the "happiness"? And how does this relate to different brain wirings of different people? Would you say per brain wiring there's a different specific ideal formula one would have to follow? E.g. there's some base things like getting proper sleep, exercise, diet, but then there's some more specific things for…

>Would you think there's a perfect formula

The advice comes from coursework most medical students must do in year 2. It is backed by research data on integrated healthcare programs.

Simply avoiding misery is not the whole equation. =3

https://www.youtube.com/watch?v=LO1mTELoj6o

Re: The Exhaustion of Talking to a Tool

#54

Post by akramachamarei »

iwontberude wrote: post #17
Jonovono wrote: post #15

Man, I don't know. LLMs (with the good and bad) feel like the first time a tool has genuinely been an extension of me. And quite the opposite. I'm a quite introverted person. Spending time in a meeting or talking with other humans I find quite exhausting. I don't really get that at all with talking with LLMs.

It’s because large language models give a variable reward and it’s essentially an addiction that many people have not even recognized.

> it’s essentially an addiction that many people have not even recognized

A little patronizing, no? Maybe some people actually like it, even if you or I don't.

Re: The Exhaustion of Talking to a Tool

#55

Post by mrandish »

I didn't expect LLMs to change their output so much depending on how I talk to them. I've done my own controlled tests and verified tone impacts quality even when content is held constant. This effect is called "Linguistic Convergence" or "Conversational Mirroring" and it's been studied extensively. The effect is minimal in coding contexts but becomes much more pronounced in collaborative contexts like creative brainstorming and concept development.

I noticed this because I use LLMs quite often as a note taker, research assistant and reference collector when I'm doing ideation, domain mapping and knowledge acquisition. In those kinds of sessions, if I issue purely directive instructions, the LLM's output quality will begin to drop quite quickly. And if adopt a tone of conversational engagement, the LLMs output quality remains high. As the article states, this can be a burdensome distraction which creates additional cognitive load. It's basically a non-economic cost to using LLMs in these contexts.

Reading research on this, it's fundamental to the nature of LLMs and can't simply be prompted away or easily fixed in fine-tuning. It's an artifact of attention dilution and contextual satiation. If I include semantic richness, structural variety, domain-specific terminology and explicit reasoning steps, it provides higher-entropy tokens to the model's attention mechanism which shifts the weight calculations toward richer areas of the model's latent space. By ingesting a composite of human's collective linguistic structures, it seems like models inherited some of our quirks and sensitivities too.

Re: The Exhaustion of Talking to a Tool

#56

Post by stillpointlab »

vladms wrote: post #43
stillpointlab wrote: post #23

I feel on the other side of this. Just yesterday I was reviewing some code output from Claude and I realized a change that I had asked for in a previous review step wasn't what I wanted. I had a moment of social anxiety, like I didn't want to bother a coworker with my indecision. But I have to remember, the LLM doesn't care. It doesn't have an ego. It doesn't get annoyed at being asked to redo work. I still say "plea…

Sorry to hear about the moment of social anxiety (I assume it happens with humans occasionally). But can't wonder don't you have also moments of joy because of job well done, an appreciative colleague or something similar? I don't like either the "negative" part, but I find it necessary to have both negatives and positives in life to create bonds, meaning and more simply, not to get bored. I would be worried that if…

I'm pointing out how I noticed a particular emotional response when working with LLMs.

I've been an engineering manager in the past and I have tried my best to keep the needs of my team in mind when I am delegating work. I try to consider the person, their goals, motivations, preferences, frustrations. I consider before interrupting them if the minor issue I am bringing up is worth the distraction it might cause them, since switching tasks is a mental load.

But with LLMs, almost none of that matters. They don't have goals, motivations or preferences in the same way people do. I can interrupt it all day and it won't get frustrated or lose motivation.

I think anxiety is a harsher word than I mean, but it is close to the feeling I have when I'm about to deliver bad news to someone. When I'm about to say "you know all that work I asked you to do, I need you to throw it away and restart". And I model in my mind the frustration and demotivation this can cause a person. And then I feel anxious about causing them this frustration.

I have to train myself out of that when instructing LLMs. It doesn't mean I have to avoid moments of joy or appreciation. It means I have to understand LLMs have different needs than people, and I have to work towards those needs.

Re: The Exhaustion of Talking to a Tool

#57

Post by mewpmewp2 »

Joel_Mckay wrote: post #53
mewpmewp2 wrote: post #50

Earlier quoted context omitted.

I think happiness is very complicated. Would you think there's a perfect formula to follow that will yield you the "happiness"? And how does this relate to different brain wirings of different people? Would you say per brain wiring there's a different specific ideal formula one would have to follow? E.g. there's some base things like getting proper sleep, exercise, diet, but then there's some more specific things for…

>Would you think there's a perfect formula The advice comes from coursework most medical students must do in year 2. It is backed by research data on integrated healthcare programs. Simply avoiding misery is not the whole equation. =3 https://www.youtube.com/watch?v=LO1mTELoj6o

Would you say you are happy as in you have solved happiness for yourself?

Re: The Exhaustion of Talking to a Tool

#58

Post by kasey_junk »

blucollar_coder wrote: post #24
kasey_junk wrote: post #13

Most of my workflows have slowly moved away from the chat interface with llms. Instead they look more like traditional Unix pipelines that just happen to call Unix tools that interact with llms. This allows me to make more repeatable processes, not be tied down to vendor implementations of workflows and mix and match models for cost and efficacy. There is nothing that ties you to talking with the text generator black…

Would you provide some examples? It sounds like you're feeding results from one LLM into another LLM invocation. Kind of like the loop thing everyone's talking about, but more like a workflow. Or programming with LLMs

In the simplest incarnation I’m just using the built into the agent cli parm’s that trigger non-tui behavior (for instance calling codex exec instead of just codex).

Re: The Exhaustion of Talking to a Tool

#59

Post by Joel_Mckay »

mewpmewp2 wrote: post #57
Joel_Mckay wrote: post #53

Earlier quoted context omitted.

>Would you think there's a perfect formula The advice comes from coursework most medical students must do in year 2. It is backed by research data on integrated healthcare programs. Simply avoiding misery is not the whole equation. =3 https://www.youtube.com/watch?v=LO1mTELoj6o

Would you say you are happy as in you have solved happiness for yourself?

On occasion, as I often chose wisdom, logic, and indifferentism. =3

Re: The Exhaustion of Talking to a Tool

#60

Post by mewpmewp2 »

Joel_Mckay wrote: post #59
mewpmewp2 wrote: post #57

Earlier quoted context omitted.

Would you say you are happy as in you have solved happiness for yourself?

On occasion, as I often chose wisdom, logic, and indifferentism. =3

Do you mean those three things are what let you solve it, or that they're what kept you from solving it more often?
Post reply on HN