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

I'd rather read the prompt

claytonwramsey.com

Post reply on HN
Print view View original

191–200 of 880 posts

Re: I'd rather read the prompt

#191

Post by palata »

_qswe wrote: post #169
palata wrote: post #130

Earlier quoted context omitted.

> the exact inverse of how I most often use AI, which is to throw a ton of information at it in a large prompt It sounds to me that you don't make the effort to absorb the information. You cherry-pick stuff that pops in your head or that you find online, throw that into an LLM and let it convince you that it created something sound. To me it confirms what the article says: it's not worth reading what you produce this…

You're making a weirdly uncharitable assumption. I'm referring to information which I largely or entirely wrote myself, or which I otherwise have proprietary access to, not which I randomly cherry-picked from scattershot Google results. Synthesizing large amounts of information into smaller more focused outputs is something LLMs happen to excel at. Doing the exact same work more slowly by hand just to prove a point t…

> Doing the exact same work more slowly by hand just to prove a point to someone on HN isn't a productive way to deliver business value.

You prove my point again: it's not "just to prove a point". It's about internalising the information, improving your ability to synthesise and be critical.

Sure, if your only objective is to "deliver business value", maybe you make more money by being uninteresting with an LLM. My point is that if you get good at doing all that without an LLM, then you become a more interesting person. You will be able to have an actual discussion with a real human and be interesting.

Re: I'd rather read the prompt

#192

Post by YmiYugy »

I mostly use LLMs as a more convenient Google and to automate annoying code transformations with a conveniency of a natural language interface. Sometimes, I use it to "improve" my writing style.

I have to admit I was a bit surprised how bad LLMs are at the continue this essay task. When I read it in the blog I suspected this might have been a problem with the prompt or the using one of the smaller variants of Gemini. So I tried it with Gemini 2.5 Pro and iterated quite a bit providing generic feedback without offering solutions. I could not get the model to form a coherent well reasoned argument. Maybe I need to recalibrate my expectations of what LLMs are capable, but I also suspect that current models have heavy guardrails, use a low temperature and have been specifically tuned for problem solving and avoid hallucinations as much as possible.

Re: I'd rather read the prompt

#193

Post by Workaccount2 »

I just want to point out that AI generated material is naturally a confirmation bias machine. When the output is obviously AI, you confirm that you can easily spot AI output. When the output is human-level, you just pass through it without a second thought. There is almost no regular scenario where you are retroactively made aware something is AI.

Re: I'd rather read the prompt

#194

Post by Wowfunhappy »

throwawaysleep wrote: post #125
jsheard wrote: post #74

I'm far from the first to make this observation but LLMs are like anti-compression algorithms when used like that, a simple idea gets expanded into a bloated mess by an LLM, then sent to someone else who runs it through another LLM to summarize it back to something approximating the original prompt. Nobody benefits aside from Sam Altman and co, who get to pocket a cool $0.000000001 for enabling this pointless exercis…

Depends on what you are looking for. I’ve turned half baked ideas into white papers for plenty of praise. I’ve used them to make my Jira tickets seem complicated and complete. I’ve used them to get praised for writing comprehensive documentation. Part of my performance review is indirectly using bloat to seem sophisticated and thorough.

...thinking about it, there are probably situations where making something more verbose makes it take less effort to read. I can see how an LLM might be useful in that situation.

Re: I'd rather read the prompt

#195

Post by SoftTalker »

lionkor wrote: post #68
oncallthrow wrote: post #37

LLM cheating detection is an interesting case of the toupee fallacy. The most obvious ChatGPT cheating, like that mentioned in this article, is pretty easy to detect. However, a decent cheater will quickly discover ways to conduce their LLM into producing text that is very difficult to detect. I think if I was in the teaching profession I'd just leave, to be honest. The joy of reviewing student work will inevitably b…

You assume that the teachers job is to catch when someone is cheating; its not. The teachers job is to teach, and if the kids don't learn because their parents allow them to cheat, don't check them at all, and let them behave like shitheads, then the kids will fail in life.

In many current-day school systems, the teachers job is to get the required percentage of students to pass the state assessment for their grade level.

They don’t get an exemption if the parents don’t care.

Re: I'd rather read the prompt

#196

Post by chmod775 »

musicale wrote: post #162
throwawaysleep wrote: post #125

Earlier quoted context omitted.

Depends on what you are looking for. I’ve turned half baked ideas into white papers for plenty of praise. I’ve used them to make my Jira tickets seem complicated and complete. I’ve used them to get praised for writing comprehensive documentation. Part of my performance review is indirectly using bloat to seem sophisticated and thorough.

> comprehensive documentation Documentation is an interesting use case. There are various kinds of documentation (reference, tutorial, architecture, etc.) and LLMs might be useful for things like - repetitive formatting and summarization of APIs for reference - tutorials which repeat the same information verbosely in an additive, logical sequence (though probably a human would be better) - sample code (though human-w…

Yes, but none of the current LLMs are even remotely useful doing that kind of work for even something moderately complex. I have a 2k LOC project that no LLM even "understands" *. They can't grasp what it is: It's a mostly react-compatible implementation of "hooks" to be used for a non-DOM application. Every code assistant thinks it's a project using React.

Any documentation they write at best re-states what is immediately obvious from the surrounding code (Useless: I need to explain why), or is some hallucination trying to pretend it's a React app.

To their credit they've slowly gotten better now that a lot of documentation already exists, but that was me doing the work for them. What I needed them to do was understand the project from existing code, then write documentation for me.

Though I guess once we're at the point AI is that good, we don't need to write any documentation anymore, since every dev can just generate it for themselves with their favorite AI and in the way they prefer to consume it.

* They'll pretend they understand by re-stating what is written in the README, then proceed to produce nonsense.

Re: I'd rather read the prompt

#197

Post by duskwuff »

ineptech wrote: post #86

Relatedly, there was a major controversy at work recently over the propriety of adding something like this to a lengthy email discussion: > Since this is a long thread and we're including a wider audience, I thought I'd add Copilot's summary... Someone called them out for it, several others defended it. It was brought up in one team's retro and the opinions were divided and very contentious, ranging from, "the summar…

LLMs aren't even that good at summarizing poorly structured text, like email discussions. They can certainly cherry-pick bits and pieces and make a guess at the overall topic, but my experience has been that they're poor at identifying what's most salient. They get particularly confused when the input is internally inconsistent, like when participants on a mailing list disagree about a topic or submit competing proposals.

Re: I'd rather read the prompt

#198

Post by staunton »

Herring wrote: post #160
staunton wrote: post #140

Earlier quoted context omitted.

I'm sure it's beside the point but... Your benchmark for "long flowing beautiful content" is apple.com? It's competing with Hemingway?

Spoken like a true "just the facts" guy. If this is a competition then Hemmingway is losing hands down. Serious well-funded marketing departments came up with that long-form style, because that's what the vast majority of people respond to. I'm just happy they have a tiny "tech specs" section for me (and presumably you).

I just went on apple.com and tried for 8 minutes to find "long beautiful flowing content". Most of the actual text I came across there were not-so-beautiful legal disclaimers...

Can you share a link to what you mean?

Re: I'd rather read the prompt

#199

Post by ponector »

necovek wrote: post #54

I've already asked a number of colleagues at work producing insane amount of gibberish with LLMs to just pass me the prompt instead: if LLM can produce verbose text with limited input, I just need that concise input too (the rest is simply made up crap).

Chatgpt very useful for adding softness and politeness to my sentences. Would you like more straight forward text which probably will be rude for regular american?

Re: I'd rather read the prompt

#200

Post by galangalalgol »

musicale wrote: post #162
throwawaysleep wrote: post #125

Earlier quoted context omitted.

Depends on what you are looking for. I’ve turned half baked ideas into white papers for plenty of praise. I’ve used them to make my Jira tickets seem complicated and complete. I’ve used them to get praised for writing comprehensive documentation. Part of my performance review is indirectly using bloat to seem sophisticated and thorough.

> comprehensive documentation Documentation is an interesting use case. There are various kinds of documentation (reference, tutorial, architecture, etc.) and LLMs might be useful for things like - repetitive formatting and summarization of APIs for reference - tutorials which repeat the same information verbosely in an additive, logical sequence (though probably a human would be better) - sample code (though human-w…

Delivering a library with an llm to explain the api and idiomatic usage seems like an interesting use case.
Post reply on HN