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

AI;DR (AI; Didn't Read)

rickmanelius.com

Post reply on HN
Print view View original

711–720 of 729 posts

Re: AI;DR (AI; Didn't Read)

#711

Post by tpoacher »

mediaman wrote: post #188
tpoacher wrote: post #129

Earlier quoted context omitted.

I completely agree in principle, but to play devil's advocate, if an "author's" prompting technique is better than mine such that they generate a more useful article than what my own AI session might produce on average, then maybe there's value in some generated articles. The above still implies some effort going into good prompting, however. I completely agree about the uselessness of "build me a castle; no bugs plz…

It's possible, but I've experimented with different style guides and prompt methods and have never gotten it to output what I'd call "good" writing. It is almost always low information density. Sometimes I can get it to remove most of the extraordinarily cliche phrases it injects into everything, along with the tortured, gauche prose (especially Claude). But what remains is an expansion algorithm: it takes a simple i…

Fair enough, but to be fair, that just sounds like bad defaults to me. Most of the things you describe mostly go away with a nice customised system prompt.

Re: AI;DR (AI; Didn't Read)

#712

Post by davidguetta »

LPisGood wrote: post #56

My coworkers continue to dump hundreds of lines of AI documentation in every PR and every other line of code has between one and ten lines of AI generated comments, talking about the real unlock and how things are byte for byte identical on the load bearing path or how the acceptance ladder is misleading. Features are coming out and metrics are improving, but we’re basically in a post readability code base, with the…

At my work i put these simple rules

- PR descriptions should be human-written, you are a human communicating to another human. it should be limited at 4 bullet points, and maybe a "details" section in the (rare) needed cases

- in general, if you want human attention, produce human effort

Re: AI;DR (AI; Didn't Read)

#713

Post by supermatt »

voxelghost wrote: post #704
supermatt wrote: post #481

Earlier quoted context omitted.

Thank you. Someone else posted similar with ChatGPT and it looks like that strategy could be a way for me to manually edit a lot more easily.

Just wanted to add; > Iff you dismiss AI writings just because it sounds like it’s written by aI you will end up missing out on some truly valuable content. Yes that might be the case. But with Ai now being in so heavy use, there is so much content generated that most of us dont have time to just read output of someone elses basic prompts, if we can't determine how much original thought went into it. So, with your Ai…

Sure, the problem is that people will just use humanisers if they aren't already.

I don't use one because my aim is to intentionally produce the obviously LLM-sounding output as that is what I find easier to read (although as others have said, I can preserve more of my "voice", if I tweak the prompt somewhat).

Re: AI;DR (AI; Didn't Read)

#714

Post by queenkjuul »

1vuio0pswjnm7 wrote: post #701

If people are comfortable reading output from their own use of "AI" then how do we explain why they are uncomfortable with output from someone's else's use of "AI" It would make sense, for example, if these people who are uncomfortable with output from others' use of "AI" are people who do not themselves use "AI"

I hate reading the output of "my AI," it's miserable verbose jargon laden slop. I tolerate reading it because if I prompted it, i either asked for code, pasted in a stack trace, or am looking for an answer to a question i failed to find via other means.

The last thing in the world that I want is to read an entire article written by AI, nor am i happy when i ask a human a question and they don't answer but instead just copy paste to Claude and back as if that's comparable

Re: AI;DR (AI; Didn't Read)

#715

Post by spopejoy »

afr0ck wrote: post #107

I think the main reason many people (including me), very often, lack the motivation to read content that is likely generated by AI is the suspicion that it comes from a place of intellectual laziness. Another reason, based on personal experience, is that AI content may suffer from too much verbosity, too much jargon and over-confidence, which makes the reading experience feel fake and border-line irritating. In many…

I struggle to read long code docs even if they're 100% human-written and maintained, mainly because the longer the text, the greater opportunities for miscommunication, staleness, repetition etc.

My point being, docs were already an unsolved problem in code. Of course LLMs dumping reams of docs is not good, but are we really sure it's worse than before? Undocumented code is bad, and mis-documented code is the worst.

Re: AI;DR (AI; Didn't Read)

#716

Post by zingababba »

atomicnumber3 wrote: post #489
stronglikedan wrote: post #248

Earlier quoted context omitted.

Doesn't matter. You'll come off as the baddie. It's the foreseeable future. Best just to learn how to interpret AI generated shit, or learn how to run it through AI and have it translate it to a more concise format without all the buzzwords. It'll take getting used to, but it'll save your career.

The number one rule of corpo programming: never, ever solve a problem you haven't been asked to solve.

Lol, yup.

Re: AI;DR (AI; Didn't Read)

#717

Post by tanseydavid »

roarcher wrote: post #519
doix wrote: post #510

Earlier quoted context omitted.

I hate that I have to change the way I write to avoid AI-isms. I loved using "load bearing" to describe weird code that you think you can delete but is actually holding everything together. Now people think I'm just parroting what Claude said. It sucks. I want my catch phrases back, I guess this is how em dash users felt

As a former em-dash enjoyer, I indeed feel your pain.

I will continue to use em-dashes as I always have.

When someone uses the presence of em-dash as the only signal for AI-generated then they are simply lazy.

Re: AI;DR (AI; Didn't Read)

#718

Post by NateEag »

a2ff6eeb0 wrote: post #644
NateEag wrote: post #497

Earlier quoted context omitted.

Your boss can get Claude to make unreadable, noisy, confusing code just as well as you can. There may not be a career left in software development with these monstrosities, but if there is, it's in using decades of experience to get the abominations to produce something other than unreadable dreck. If the code itself truly doesn't matter any more, programmers will no longer be paid.

Oh, yeah, the amount of skill needed for programming is quite low these days, the people paying just haven't figured it out quite yet. There's still some needed, but it's closer to line cook than nuclear physicist. But the managers won't do it, because you still need manual testing to make sure that the AI models got things right. Managers will want to hire someone to do the manual testing.

The amount of skill required is still pretty high, IMO - if we set a requirement that a programmer thoroughly understand what they're shipping.

If we remove that requirement (as most around me seem to be doing), then yeah, not much skill required, but the system is going to steadily trend towards being a buggy, unmaintainable pile of sloppypasta.

Maybe the models get good enough they can handle maintaining that, but anecdotal reports so far seem to indicate we aren't there yet.

I guess we'll see if AI 2027 pans out.

Re: AI;DR (AI; Didn't Read)

#719

Post by mannanj »

annzabelle wrote: post #695
wpietri wrote: post #626

Earlier quoted context omitted.

Oh wow, that sounds horrible. Do you have a sense of what motivates that behavior?

The founder CEO (with a cult of personality) really believes that it is the way to "top tech talent." He comes from a management consulting background and tries to manage the software organization like they're consultants. Leads to a very bottom heavy organization, with a lot of juniors, many of whom get PIPed right as they're starting to become useful. It leads to high turnover with a lot of backstabbing, and very s…

Ah yeah that explains my experience a bit better. I was there in the COVID wfh time, 2020 or so, then rejoined 2024 when I experienced things not too unlike what you described. (except I joined as a principal, ~8-10 years of experience, and received backstabbing and lies from juniors and managers (edit and tech lead)).

I don't recommend going to Plano, TX and joining the Auto Navigator product division.

Re: AI;DR (AI; Didn't Read)

#720

Post by ninkendo »

epidemian wrote: post #524
ninkendo wrote: post #389

Earlier quoted context omitted.

> Document what's there, not the diff We recently added a similar thing to our style guide, It’s astonishing to me that we have to spell this out, that something as obvious as this needs to be explained to LLM’s at all. They’re supposed to be exceeding human intelligence, at least at things like programming, but can’t understand basic things like what code comments are.

> It’s astonishing to me that we have to spell this out, that something as obvious as this needs to be explained to LLM’s at all. Hehe. Yeah, that tendency of LLMs to document "the story" of the code instead of its current purpose (or non-obvious implementation details) is a pet peeve of mine too. I've added a slew of guidelines to try to sway Claude to not do this, but it still does it often. At the same time, it fe…

RLHF has the same problem as human reviews of AI code: AI code (and comments) look plausible at first, and if you have 100 other PR's to get to, it looks "good enough" and you approve it. I'm sure the humans doing the "human feedback" part of RLHF at anthropic are just as tired as I am at reading all of it, and start to just approve it when it looks plausible.

It's doubly insidious because it trips up the human brain too: When I'm reading a PR saying "fix lock ordering to avoid deadlocks on user deletion", and there's a comment somewhere in the diff saying "// use the fixed lock ordering here", my brain tends to completely forget the fact that the comment doesn't make any sense in its surrounding context. Because it makes perfect sense in the context of being the human reviewing the diff. But it's a slight bit of mental effort to remind yourself "what is this comment going to look like to someone reading the code after this is merged?"

I wouldn't be surprised whatsoever if the RLHF supervisors forget to apply that extra bit of mental effort and say "yup, this comment looks great", forgetting to check the surrounding code to see if it makes sense on its own.

Post reply on HN