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

It's insulting to read AI-generated blog posts

blog.pabloecortez.com

Post reply on HN
Print view View original

421–430 of 565 posts

Re: It's insulting to read AI-generated blog posts

#421

Post by array_key_first »

dingnuts wrote: post #396
oceanplexian wrote: post #350

Earlier quoted context omitted.

LLMs write things in a certain style because that's how the base models are fine tuned before being given to the public. It's not because they can't write PRs indistinguishable from humans, or can't write code without Emojis. It's because they don't want to freak out the general public so they have essentially poisoned the models to stave off regulation a little bit longer.

this is WILD speculation without a citation. it would be a fascinating comment if you had one! but without? sounds like bullshit to me...

It is wildly speculative, but it's something I've never considered. If I were making a brave new technology that I knew had power for unprecedented evil, I might gimp it, too.

Re: It's insulting to read AI-generated blog posts

#422

Post by array_key_first »

abixb wrote: post #18
icapybara wrote: post #9

If they can’t be bothered to write it, why should I be bothered to read it?

I'm sure lots of "readers" of such articles fed it to another AI model to summarize it, thereby completely bypassing the usual human experience of writing and then careful (and critical) reading and parsing of the article text. I weep for the future. Also, reminds me of this cartoon from March 2023. [0] [0] https://marketoonist.com/2023/03/ai-written-ai-read.html

Are people doing this or is this just what, like, Apple or someone is telling us people are doing?

Because I've never seen anyone actually use a summarizing AI willingly. And especially not for blogs and other discretionary activities.

That's like getting the remote from the hit blockbuster "Click" starring Adam Sandler (2006) and then using it to skip sex. Just doesn't make any sense.

Re: It's insulting to read AI-generated blog posts

#423

Post by strken »

JohnFen wrote: post #326
ok_dad wrote: post #211

Earlier quoted context omitted.

> Code review is one of the places where experience is transferred. Almost nobody uses it for that today, unfortunately, and code reviews in both directions are probably where the vast majority of learning software development comes from. I learned nearly zilch in my first 5 years as a software dev at crappy startups, then I learned more about software development in 6 months when a new team actually took the time to…

I agree. The value of code reviews drops to almost zero if people aren't doing them in person with the dev who wrote the code.

I disagree. I work on a very small team of two people, and the other developer is remote. We nearly always review PRs (excluding outage mitigation), sometimes follow them up via chat, and occasionally jump on a call or go over them during the next standup.

Firstly, we get important benefits even when there's nothing to talk about: we get to see what the other person is working on, which stops us getting siloed or working alone. Secondly, we do leave useful feedback and often link to full articles explaining concepts, and this can be a good enough explanation for the PR author to just make the requested change. Thirdly, we escalate things to in-person discussion when appropriate, so we end up having the most valuable discussions anyway, which are around architecture, ongoing code style changes, and teaching/learning new things.

I don't understand how someone could think that async code review has almost zero value unless they worked somewhere with a culture of almost zero effort code reviews.

Re: It's insulting to read AI-generated blog posts

#424

Post by falcor84 »

falcor84 wrote: post #126
latexr wrote: post #114

Earlier quoted context omitted.

This makes no sense, and it’s absurd anyone thinks it does. If the AI PR were any good, it wouldn’t need review. And if it does need review, why would the AI be trustworthy if it did a poor job the first time? This is like reviewing your own PRs, it completely defeats the purpose. And no, using different models doesn’t fix the issue. That’s just adding several layers of stupid on top of each other and praying that so…

> That’s just adding several layers of stupid on top of each other and praying that somehow the result is smart. That is literally how civilization works.

Just to explain my brusque comment: the way I see it, civilization is populated with a large fraction of individuals whose intelligence or conscientiousness I wouldn't trust to mind my cactus, but that I'm ok with entrusting a lot more too because of the systems and processes offered by society at large.

As an example, knowing that a service is offered by a registered company with presence in my area gives me the knowledge "that they know that I know" that if something goes wrong, I can sue them for negligence, possibly up to piercing the corporate veil the company and having the directors serve prison time. From that I can somewhat rationally derive that if the company has been in business offer similar services for years, it is likely that they have processes in place to maintain a level of professionalism that would lower the risk of such lawsuits. And on an organisational level, even if I still have good reason to think that most of the employees are incompetent, the fact that the company is making it work gives me a significantly higher preference in the "result" than I would in any individual "stupid" component.

And for a closer-to-home example, the internet is well known to be a highly reliable system built from unreliable components.

Re: It's insulting to read AI-generated blog posts

#425

Post by godelski »

mikepurvis wrote: post #417
godelski wrote: post #272

Earlier quoted context omitted.

> But once that’s done, why not? Do you have the same understanding of the code? Be honest here. I don't think you do. Just like none of us have the same understanding of the code somebody else wrote. It's just a fact that you understand the code you wrote better than code you didn't. I'm not saying you don't understand the code, that's different. But there's a deeper understanding to code you wrote, right? You might…

So the most recent thing that I did a bunch of vibe coding on was typescript actions for GHA. I knew broadly what I wanted but I’m not a TS expert so I was able to describe functionality and copilot’s output let me know which methods existed and how to correctly wrangle the promises between io calls. It undoubtedly saved me time vs learning all that first, and in fact was itself a good chance to “review” some decent…

I think this is a different use case. The context we're talking about is building software. A GitHub action is really a script. Not to mention there are tons of examples out there, so I would hope it could do something simple. Vibe coding scripts isn't what people are typically concerned about.

  > but I’m not a TS expert
Although this is ultimately related. How can you verify that it is working as intended? You admit to not having those skills. To clarify, I'm sure "it's working" but can you verify the "as intended" part? This is the hard part of any coding. Getting things working isn't trivial, but getting things working right takes a lot more time.

  > So the most recent thing that I did
I'll share a recent thing I tried too...

I was working on a setup.py file and I knew I had done something small and dumb, but was being blind to it. So I pulled up claude code and had it run parallel to my hunt. Asked it to run the build command and search for the error. It got caught up in some cmake flags I was passing, erroneously calling them errors. I get a number of prompts in and they're all wrong. I fixed the code btw, it was a variable naming error (classic!).

I've also had success with claude, but it is super hit or miss. I've never gotten it to work well for anything remotely complicated if there also isn't the code in a popular repo I could just copy paste. But it is pretty hit or miss for even scripts, which I write a lot of bash. People keep telling me it is great for bash and honestly guys, just read the man pages... (and use some god damn functions!)

Re: It's insulting to read AI-generated blog posts

#426

Post by cyrusradfar »

jihadjihad wrote: post #23

It's similarly insulting to read your AI-generated pull request. If I see another "dart-on-target" emoji... You're telling me I need to use 100% of my brain, reasoning power, and time to go over your code, but you didn't feel the need to hold yourself to the same standard?

It seems I'm going to be contrarian here because I really prefer AI-supported (obviously reviewed for accuracy) PR comments over what I had seen before where I'd, often, need to reach out to someone to ask follow up questions on requirements, a link to a ticket, or any number of omissions.

I have worked at smaller firms, mostly, early stage (In the end, I'll have to say "it depends" -- you can't just throw slop at people but there's definitely a middle ground where everyone wins.

Re: It's insulting to read AI-generated blog posts

#427

Post by falcor84 »

latexr wrote: post #225
px43 wrote: post #212

Earlier quoted context omitted.

> If the AI PR were any good, it wouldn’t need review. So, your minimum bar for a useful AI is that it must always be perfect and a far better programmer than any human that has ever lived? Coding agents are basically interns. They make stupid mistakes, but even if they're doing things 95% correctly, then they're still adding a ton of value to the dev process. Human reviewers can use AI tools to quickly sniff out com…

> So, your minimum bar for a useful AI is that it must always be perfect and a far better programmer than any human that has ever lived? You are transparently engaging in bad faith by purposefully straw manning the argument. No one is arguing for “far better programmer than any human that has ever lived”. That is an exaggeration used to force the other person to reframe their argument within its already obvious conte…

I strongly disagree that it was bad faith or strawmanning. The ancestor comment had:

> This makes no sense, and it’s absurd anyone thinks it does. If the AI PR were any good, it wouldn’t need review. And if it does need review, why would the AI be trustworthy if it did a poor job the first time?

This is an entirely unfair expectation. Even the best human SWEs create PRs with significant issues - it's absurd by the parent to say that if a PR is "any good, it wouldn’t need review"; it's just an unreasonable bar, and I think that @latexr was entirely justified in pushing back against that expectation.

As for the "95% correctly", this appears to be a strawman argument on your end, as they said "even if ...", rather than claiming that this is the situation at the moment. But having said that, I would actually like to ask both of you - what does it even mean for a PR to be 95% correct - does it mean that that 95% of the LoC are bug-free, or do you have something else in mind?

Re: It's insulting to read AI-generated blog posts

#428

Post by godelski »

jmcodes wrote: post #393
D13Fd wrote: post #381

Earlier quoted context omitted.

If you wrote the code, then you’ll understand it and know why it is written the way you wrote it. If the AI writes the code, you can still understand the code, but you will never know why the code is written that way. The AI itself doesn’t know, beyond the fact that that’s how it is in the training data (and that’s true even if it could generate a plausible answer for why, if you asked it).

I don't agree entirely with this. I know why the LLM wrote the code that way. Because I told it to and _I_ know why I want the code that way. If people are letting the LLM decide how the code will be written then I think they're using them wrong and yes 100% they won't understand the code as well as if they had written it by hand. LLMs are just good pattern matchers and can spit out text faster than humans, so that's…

  > I know why the LLM wrote the code that way. Because I told it to and _I_ know why I want the code that way.
That's a different "why".

  > If people are letting the LLM decide how the code will be written then I think they're using them wrong
I'm unconvinced you can have an LLM produce code and you do all the decision making. These are fundamentally at odds. I am convinced that it will tend to follow your general direction, but when you write the code you're not just writing either.

I don't actually ever feel like the LLMs help me generate code faster because when writing I am also designing. It doesn't take much brain power to make my fingers move. They are a lot slower than my brain. Hell, I can talk and type at the same time, and it isn't like this is an uncommon feat. But I also can't talk and type if I'm working on the hard part of the code because I'm not just writing.

People often tell me they use LLMs to do boilerplate. I can understand this, but at the same time it begs the question "why are you writing boilerplate?" or "why are you writing so much boilerplate?" If it is boilerplate, why not generate it through scripts or libraries? Those have a lot of additional benefits. Saves you time, saves your coworkers time, and can make the code a lot cleaner because you're now explicitly saying "this is a routine". I mean... that's what functions are for, right? I find this has more value and saves more time in the long run than getting the LLMs to keep churning out boilerplate. It also makes things easier to debug because you have far fewer things to look at.

Re: It's insulting to read AI-generated blog posts

#430

Post by nativeit »

freedomben wrote: post #405
WesolyKubeczek wrote: post #388

Earlier quoted context omitted.

It’s below me to use LLMs to comment on HN.

Exactly what an LLM would say. Jk, your comments don't seem at all to me like AI. I don't see how that could even be suggested

[flagged]
Post reply on HN