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

361–370 of 729 posts

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

#361

Post by jdidjsjdjw »

Insimwytim wrote: post #243
gortok wrote: post #36

The part that astonishes me is that in the year of our common era two thousand twenty-six that it's not universally offensive and reviling to post an AI-generated response to another person. If I'm reading something on the internet, I'm either reading it to learn, or I'm reading it to be persuaded. If I wanted the LLM to teach me (thank you, no), I would ask an LLM. I'm reading your website/newsletter/email because I…

I think there are some, who couldn't be a part of the public conversation, precisely because of they inability to add anything to said conversation. LLMs enables them to post (or, rather, paste) something, in an attempt to trick others into believing they do have something to say. They think they sound smart now, because LLMs are "very much the “stupid person’s idea of an intelligent person”"[1] So now we are flooded…

> So now we are flooded with stupid people's attempts to show us how smart they are.

Now? My friend, we’ve been flooded with stupid people’s attempts to show us how smart they are ever since internet forums came to be. Heck, just look at 90% of the comments on any given HN post. It’s mind-numbingly stupidity by the handful disguised as PhD talk—big words, fancy terms, heavy phrasing, all to say a bunch of nothing or just pretend they have a part in said conversation.

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

#362

Post by JimTheMan »

zirkonit wrote: post #85
sega_sai wrote: post #5

If some text is AI written as a response to a much shorter prompt, then the prompt and/or sources used to make the text should be published instead (or at the very least together with the text).

Funnily enough, for me AI writing is usually the summarizer rather than expander. I would usually give Claude 2-3x the amount of data in notes, context, braindump etc; when doing person-to-person communication, using AI to _expand_ the context rather than shrink it is doing everyone a disservice.

The problems with AI text is mostly a 'skill issue' or 'taste issue'.

It's a tool like any other. It assists the intelligent, but makes the lazy - lazi-er.

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

#363

Post by mc3301 »

jedberg wrote: post #285

Every time this comes up I have to ask, how do you know it is AI? Sure, sometimes there are strong tells, like if they forget to take off the "Do you want me to change anything" at the end. But AI was trained on good writing. Good writing looks like AI. I've had to reduce my use of em dashes for example because they are an "AI tell". I used to like using the term "load bearing". And the list gets longer every day. Ev…

[flagged]

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

#364

Post by jedberg »

asadotzler wrote: post #357
jedberg wrote: post #297

Earlier quoted context omitted.

Sure, that's one of those obvious tells. But when I sit and spend 10 minutes making a well thought out and researched comment, and then properly format it in markdown, and then the first response is "this is AI", it's really frustrating.

If you read more literature you will get better at writing and your writing should stop being as similar to the banal mess that AI spits out. This is on you to differentiate.

I’ve read plenty of literature. So has the AI training. That won’t help. In fact, it will make it worse.

I even studied writing in college. I’ve written published books and articles. I can write well. But thanks for the tip!

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

#365

Post by hinkley »

This has really started to hit my radar because people are filing PRs with AI generated commit messages that read like a bill being presented to congress. Holy shit, give me a summary.

I don't know if the contributors are writing any of the code they're submitting or just using AI to hop on every 'help wanted' tagged ticket in every product database but if it looks mostly like a smart human wrote it, which usually takes two rounds of reviews, then it can probably go in.

But if you make me work for it more than you did I'm just going to close the PR and do it myself.

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

#366

Post by hinkley »

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…

I told someone this week, who (or whose AI) chose to do a problem the hard way that it's usually a bad sign if you need more comments than code to solve a problem, and then suggested a couple lines of code that accomplished the same thing and used, are you sitting down? MEANINGFUL VARIABLE NAMES to document the purpose of each calculation.

I wonder if I can get a MacArthur grant for this epiphany...

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

#367

Post by supermatt »

Planktonne wrote: post #354
supermatt wrote: post #341

Earlier quoted context omitted.

I had 3 main positions - are you disagreeing with them all? You have not directly engaged with any of them. 1) Content shouldn't be judged solely on the use of an LLM. 2) There is a distinction between slop and assisted authorship. 3) LLMs can be valuable accessibility tools. You are now saying that the OUTPUT was thoughtless, whereas before you said that "I" had not put meaningful thought into it. Those are differen…

> are you disagreeing with them all Yes. I think an LLM can be useful for trivial or functional tasks, but for expressing complex positions, it's worse than useless. It's absolutely valid to dismiss LLM-generated prose, and people who who think they're doing 'assisted authorship' tend to be mistaken. I'm not saying you are incapable of thought; I'm saying that you've abdicated that by transmitting through an LLM.

> Yes

Well you are of course entitled to disagree.

> I'm not saying you are incapable of thought; I'm saying that you've abdicated that by transmitting through an LLM.

What you actually said was that I "didn't put any meaningful thought into it". Thats the INPUT, not OUTPUT. Anyway. I think we're as far as we are going to get with this "discussion". Have a good evening!

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

#368

Post by supermatt »

jcranmer wrote: post #352
supermatt wrote: post #162

Earlier quoted context omitted.

How is this? https://claude.ai/share/99bfee0f-bab9-4593-aabb-377e71f1151d

The stream-of-consciousness is only hard to read because it lacks paragraph breaks, sometimes sentence breaks, and there's so many misspelled words. If you just tweak it to fix that stuff, it becomes fairly readable, like most internet comments or whatnot. (Which do you think this comment is, a polished essay, or lightly-formatted stream of consciousness?) Meanwhile, the LLM-generated essay is hard to read because it…

Thanks. Im actually seeing that myself since someone helped me with formatting the word salad.

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

#369

Post by hinkley »

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…

My question is always, "Am I willing to maintain this code?"

Sometimes that's a yes, even if AI generated it and I can tell despite someone making half an effort to be a human for a minute during the PR process. But sometimes it's not.

It gets really dicey when you've kept an architecture a certain shape because there's a difficult feature you intend to slot into the negative space. There are two or three ways to implement that feature, and the one they chose just stomps all through your potential. But that's why we have code reviews. You don't have to read my mind if I just tell you. But it can be frustrating if someone asks you to do it a different way, and then has comments on how you accomplished that, and then comments on the fixes for your other comments.

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

#370

Post by hinkley »

cortesoft wrote: post #48

I really like the point I read somewhere the other day; instead of sending me the AI output, just send me the prompt you used to generate it. That is the only part that contains only the information you are trying to convey. The rest is just guesses flowery language added, and it confuses the actual message being sent.

Somewhere in that wall of text is a single statement that might sink the entire project...
Post reply on HN