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

The L in "LLM" Stands for Lying

acko.net

Post reply on HN
Print view View original

371–380 of 500 posts

Re: The L in "LLM" Stands for Lying

#371

Post by mikkupikku »

Jooror wrote: post #273
mikkupikku wrote: post #117

Earlier quoted context omitted.

> I’m curious about how you landed “git gud; prompt better” and not “maybe the domain I work in is a better fit for LLM code”. 1. Personal experience. Lazy prompting vs careful prompting. 2. They're coincidentally good at things I'm good at, and shit at things I don't understand. 3. Following from 2, when used by somebody who does understand a problem space which I do not, they easily succeed. That dog vibe coding ga…

Irrespective of the domain you specifically listed in 3 (game dev is, believe it or not, one of the “more complex” domains), you have completely failed to miss the point. > 2. They're coincidentally good at things I'm good at, and shit at things I don't understand. This may well be! In the perfect world this would be balanced with the knowledge that maybe “the things you’re good at” are objectively* easier than “thin…

I have definitely considered the possibility that I'm simply good at easy things and the LLM is good at easy things, and that hard things are hard for both of us. And there certainly must be some element of that going on, but I keep noticing that different people get different quality results for the same kind of problems, and it seems to line up with how good they themselves would be at that task. If you know the problem space well, you can describe the problem (and approaches to it) with a precision that people unfamiliar with the problem space will struggle with.

I think you can observe this in action by making vague requests, seeing how it does, then roll back that work and make a more precise request using relevant jargon and compare the results. For example, I asked claude to make a system that recommends files with similar tags. It gave me a recommender that just orders files by how many tags they had in common with the query file. This is the kind of solution that somebody may think up quick but it doesn't actually work great in practice. Then I reverted all of that and instead specified that it should use a vector space model with cosine similarity. It did pretty good but there was something subtly off. That is however about the limit of my expertise in this direction, so I tabbed over to a session with ChatGPT and discussed the problem on a high level for about 20 minutes, then asked ChatGPT to write up a single terse technically precise paragraph describing the problem. I told ChatGPT to use no bullet points and write no psuedocode, telling it the coding agent was already an expert in the codebase so let it worry about the coding. I give that paragraph to claude and suddenly it clicks, it bangs out a working solution without any drama. So I conclude the quality of the prompting determined the quality of the results.

Re: The L in "LLM" Stands for Lying

#372

Post by mikkupikku »

simonask wrote: post #236
mikkupikku wrote: post #112

Earlier quoted context omitted.

I think that's true, but something even more subtle is going on. The quality of the LLM output depends on how it was prompted in a way more profound than I think most people realize. If you prompt the LLM using jargon and lingo that indicate you are already well experienced with the domain space, the LLM will rollplay an experienced developer. If you prompt it like you're a clueless PHB who's never coded, the LLM wil…

This is magical thinking. LLMs are physically incapable of generating something “well thought out”, because they are physically incapable of thinking.

I don't care if the machine has a soul, I only care what the machine can produce. With good prompting, the machine produces more ""thoughtful"" results. As an engineer, that's all I care about.

Re: The L in "LLM" Stands for Lying

#373

Post by TimTheTinker »

hwers wrote: post #84
utopiah wrote: post #81

Earlier quoted context omitted.

> to teach you things way faster than the old alternatives I'm not sure if you ever had a teacher or instructor that you didn't trust, because they were a compulsive liar or addiction or any other issue. I didn't (as least not that I can remember) but I know I would be VERY on guard about it. I imagine I would consequently be quite stressed learning with them, even if they were brilliant, kind, etc. It would feel a b…

I agree, it can be incredibly frustrating at times. My rule is that if it “compiles” in my brain as an understood idea then i accept it. I also push back a lot (sometimes it points out good errors in my thinking, sometimes it admits it hallucinated). Real humans hallucinate a lot as well or confidently state subtly wrong ideas, it’s a good habit anyway. It’s basically the same approach when presented with a “formula”…

> My rule is that if it “compiles” in my brain as an understood idea then i accept it.

Unfortunately, individual people are not anywhere as reliable as a compiler for ensuring compliance to reality. We are particularly susceptible to flattery and other emotional manipulation, which LLMs frequently employ. This becomes particularly problematic when you ask for feedback on an idea.

In that case, a useful hack is to frame prompts as if you're an impartial observer and want help evaluating something, not as if the idea under evaluation is your own.

Re: The L in "LLM" Stands for Lying

#374

Post by moffkalast »

vips7L wrote: post #274
MagicMoonlight wrote: post #221

Earlier quoted context omitted.

Before mass production, the women of the household would be forced to spend every free moment of their day, outside of their other work, making fabric. Before mechanised farming, the men were forced to spend all day in the fields. Never again.

And yet many of us would prefer to be in a field instead of behind a laptop screen all day.

Have you ever tried it? I did, and sure as hell wouldn't.

Nobody's really stopping you from studying agriculture and working in that proverbial field either.

Re: The L in "LLM" Stands for Lying

#375

Post by tovej »

WarmWash wrote: post #345
dec0dedab0de wrote: post #271

Earlier quoted context omitted.

No, it's simply untrue. Players only object against AI art assets. And only when they're painfully obvious. No one cares about how the code is written. This reminded me of a conversation about AI I had with an artist last year. She was furious and cursing and saying how awful it is for stealing from artists, but then admitted she uses it for writing descriptions and marketing posts to sell her art.

Everyone is in it for themselves. The world makes waay more sense when you really internalize that. It doesn't necessarily mean people are selfish, large groups often have aligned interests, but when an individuals interest alignment changes, then their group membership almost always changes too. I'd bet she has a bunch of pirated content and anti-copyright remarks from the golden age of piracy as well.

That's not true. Most people are interested in fostering a community, even when it means sacrifice.

There _have_ however been studies that show that this attitude is prevalent in (neoclassical) economics students and others who are exposed to (neoclassical) economic thinking: https://www.sciencedirect.com/science/article/abs/pii/S22148...

It's very effective propaganda. And we have a good example of it here. (Mot saying you're spreading it maliciously, but you are spreading it).

Re: The L in "LLM" Stands for Lying

#376

Post by miyoji »

slibhb wrote: post #306
vips7L wrote: post #297

Earlier quoted context omitted.

> I don't think this will happen Block just laid off 40% of their company citing AI.

Tech companies have been laying off employees for a while now. I think it's mostly due to pandemic overhiring and higher interest rates but I suppose we'll see.

> I think it's mostly due to pandemic overhiring and higher interest rates

It's not because of pandemic overhiring, and if that were true, the layoffs in 2021-2022 would have handled it. It's 2026. The people getting laid off (on average) haven't worked at these companies since before the pandemic, they got hired in ~2023 (average tenure at a tech company is ~3 years).

It's not because of AI either. Nobody is replacing jobs with AI, AI can't do anyone's job.

It's not because of interest rates. People hired like crazy when interest rates were this high in the oughts.

It's because Elon Musk's Twitter purchase and subsequent management convinced every executive in tech that you can cut to the bone, fuck your product's quality completely, and be totally fine. It's not true, but the downsides come later and the cash influx comes now, so they're doing it anyway.

Re: The L in "LLM" Stands for Lying

#377

Post by moffkalast »

aosnsbbz wrote: post #175
noemit wrote: post #106

Many people don't know this, but the Luddites were right. I studied Art History and this particular movement. One of the claims of the Luddites is that quality would go down, because their craft took half a lifetime to master (it was passed down from parent to chile.) I was able to feel wool scarves made in europe from the middle ages. (In museum storage, under the guidance of a curator) They are a fundamentally diff…

A big difference is cutting quality for the sake of mass production when it enables creating more necessities for people to live is a good thing. It is a good tradeoff. Cutting quality to make previously deterministic software more non deterministic does not improve anyones life except Sam Altman, Dario Amodei and the rest of the billionaire class. I have no doubt in the future there will be a class of vibe software…

Agreed. Those expensive silk scarves are worth exactly zilch to the average person if they'll never see one in their entire life. They might as well not exist.

Mass production makes things accesible, and if the handmade product cannot compete relative to it, then it's clearly not that much better or some people would still pay that premium to keep it around.

With programming that's very much the case, nobody's gonna vibe code a self driving car stack or a production grade DBMS. Even the cheapest scarf still works as a scarf in 99.9% of use cases though, if maybe not for as long.

Re: The L in "LLM" Stands for Lying

#378

Post by recursive »

mr_toad wrote: post #136
noemit wrote: post #106

Many people don't know this, but the Luddites were right. I studied Art History and this particular movement. One of the claims of the Luddites is that quality would go down, because their craft took half a lifetime to master (it was passed down from parent to chile.) I was able to feel wool scarves made in europe from the middle ages. (In museum storage, under the guidance of a curator) They are a fundamentally diff…

> One of the claims of the Luddites is that quality would go down, because their craft took half a lifetime to master (it was passed down from parent to chile.) Sounds like a tautology. If you deliberately hoard knowledge of course it’s going to be hard to obtain.

You talking about the billion dollar model training efforts done in secret?

Re: The L in "LLM" Stands for Lying

#379

Post by recursive »

artisin wrote: post #102
heavyset_go wrote: post #69

Earlier quoted context omitted.

Books are just simple theses and themes with hundreds of pages of boilerplate

Hit songs are just simple four-chord loops stretched over three minutes of synthetic boilerplate.

There are many orders of magnitude more songs based on four chord loops than there are hit songs. Some people say it's easy to make a hit song. But there are a lot more people that want to do it than those that succeed. So I say no. Your take is reductive, and there is necessarily more to it.

Re: The L in "LLM" Stands for Lying

#380

Post by rybosworld »

tovej wrote: post #349
rybosworld wrote: post #126

Earlier quoted context omitted.

When web search first arrived, the same thing happened. That is, some people didn't like using the tool because it wasn't finding what they wanted. This is still true for a lot of folks today, actually. It's less "git gud; prompt better", and more, "be able to explain (well) what you want as the output". If someone messages the IT guy and says "hey my computer is broken" - what sort of helpful information can the IT…

I can assure you I give LLMs all the information they need. Including hints to what kind of solution to use. They still fail.

So how do you rectify your anecdotal experience against those made by public figures in the industry who we can all agree are at least pretty good engineers? I think that's important because if we want to stay ~anonymous, neither you nor I can verify the reputation of one another (and therefore, one another's relative master of the "Craft").

Here are some well known names who are now saying they regularly use LLM's for development. For many of these folks, that wasn't true 1-2 years ago:

- Donald Knuth: https://www-cs-faculty.stanford.edu/%7Eknuth/papers/claude-c...

- Linus Torvalds: https://arstechnica.com/ai/2026/01/hobby-github-repo-shows-l...

- John Carmack: https://x.com/ID_AA_Carmack/status/1909311174845329874

My point being - some random guy on the internet says LLM's have never been useful for them and they only output garbage vs. some of the best engineers in the field using the same tools, and saying the exact opposite of what you are.

Post reply on HN