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

531–540 of 565 posts

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

#531

Post by kagerou74 »

alyxya wrote: post #54

I personally don’t think I care if a blog post is AI generated or not. The only thing that matters to me is the content. I use ChatGPT to learn about a variety of different things, so if someone came up with an interesting set of prompts and follow ups and shared a summary of the research ChatGPT did, it could be meaningful content to me. > No, don't use it to fix your grammar, or for translations, or for whatever el…

I agree. I remember back in the days when facebook came out around mid 2000s or so my mentor started an online open-source academic journal. that was frowned upon by his colleagues. if it wasn't printed, it wasn't considered authentic. And then he went ahead and asked me (a helper, a volunteer) to create a facebook and Twitter accounts to spread the word. and that was frowned upon, too. it was considered like a really crass move among his respectable colleagues. Now social media has become a given.

Why do I need to spend hours on formatting when AI can help clean them up in seconds? Or typos? grammar inconsistencies? Clarifying sentences? I'd rather much read writings cleaned up by an AI than some incoherent fluff that really says nothing and wasting space.

And frankly, in my experience, writing with AI is no more easier than writing on your own. It's just that my priority shifts from being distracted on those externalities (like fixing typos and formatting) to the subject matter of writing itself. So I go deep dive researching topics and contents at hand with AIs and try to figure out the best way to communicate the ideas to the audience with an AI. It takes time and it's a work of its own. It takes hours depending on what I am doing.

And you probably wished I have cleaned this comment up with an AI, too! (I tried in my best human way given time I'm allowed).

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

#533

Post by bavell »

Sharlin wrote: post #524
apwheele wrote: post #517

Earlier quoted context omitted.

I do remember 1 example of an emoji in tech docs before all of this -- learning github actions (which based on my blog happened in 2021 for me, before ChatGPT release), at one point they had an apple emoji at the final stage saying "done". (I am sure there are others, I just do not remember them.) But agree excessive emoji's, tables of things, and just being overly verbose are tells for me anymore.

I do recall emoji use getting more popular in docs and – brrh – in the outputs of CLI programs already before LLMs. I’m pretty sure thst the trend originated from the JS ecosystem.

Really hate this trend/style. Sucks that it's ossified into many AIs. Always makes me think of young preteens who just started texting/DMing. Grow up!

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

#534

Post by GuinansEyebrows »

tpmoney wrote: post #412
GuinansEyebrows wrote: post #246

Earlier quoted context omitted.

I’d rather be insulted for something I am and can at least try to improve, than praised for something I’m not or can’t do, despite my physiological shortcomings.

On the other hand, your perspective is shaped by not being dismissed by the vast majority of the people you encounter for that shortcoming. I would imagine you might feel very differently if every person you met treated you as an imbecile because you were't articulate enough, especially if your best efforts at improving don't move the needle much. I can't speak for the OP's experiences, but my early schooling years w…

> On the other hand, your perspective is shaped by not being dismissed by the vast majority of the people you encounter for that shortcoming.

not to get super personal, but that's... not the case for me. i just feel differently about it, that's all!

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

#535

Post by MisterTea »

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?

"Bruh, you're supposed to use the AI to read and vet the requests so you can spend more time arguing on the internet about the merits of using AI"

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

#536

Post by blackhaj7 »

I used to be really good at spotting it but my AI radar is getting less effective now. Often I don’t notice until late in an article which is very annoying.

There are some videos on Instagram that I didn’t notice were AI until my wife told me!

If I want AI content, I will go to an AI. The only good outcome is that I am spending way less time on social media because so much of it is now AI

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

#537

Post by GuinansEyebrows »

buu700 wrote: post #485
AppleBananaPie wrote: post #483

Earlier quoted context omitted.

In my experience if the ai voice was immediately noticeable the writing provided nothing new and most of the time is actively wrong or trying to make itself seem important and sell me on something the owner has a stake in. Not sure if this is true for other people but it's basically always a sign of something I end up wishing I hadn't wasted my time reading. It isn't inherently bad by any means but it turns out it's…

That was essentially my takeaway. The problem isn't when AI was used. It's when readers can accurately deduce that AI was used. When someone uses AI skillfully, you'll never know unless they tell you.

i feel like i've seen this comparison made before, but LLMs, when used, are best applied like autotune. 99% of vocal recordings released on major (and even indie) labels have some degree of autotune applied. when done correctly, you can't tell (unless you're a grizzled engineer who can hear 1dB of compression or slight EQ changes). it's only when it's cranked up or used lazily that it can detract from the overall product.

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

#538

Post by ComplexSystems »

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?

You're absolutely right! Here is the correct, minimal reprex to demonstrate the issue:

# Minimal Reprex (Correct)

(unintelligible nonsense here)

And here is the correct, minimal fix, guaranteed to work:

# Correct Fix (Correct)

(same unintelligible nonsense, wrapped in a try/catch block)

Make this change and your code should work perfectly!

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

#539

Post by singleshot_ »

zygentoma wrote: post #501
singleshot_ wrote: post #469

Earlier quoted context omitted.

I believe it’s .4 calories per serving which is less than one and which rounds down to zero, but it’s not approximately zero by a long shot.

How is 0.4 kcal "not approximately zero by a long shot"? Especially when compared to a standard coke with around 150 kcal.

Well, it’s almost half a calorie, to begin with.

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

#540

Post by waste_monk »

bmacho wrote: post #513
ManuelKiessling wrote: post #312

Earlier quoted context omitted.

Why have the LLMs „learned“ to write PRs (and other stuff) this way? This style was definitely not mainstream on Github (or Reddit) pre-LLMs, was it? It’s strange how AI style is so easy to spot. If LLMs just follow the style that they encountered most frequently during training, wouldn’t that mean that their style would be especially hard to spot?

For this "LLM style were already the most popular, that's how LLM works, then how come LLM style is so weird and annoying" I have 2 theories. First, LLM style did not even exist, it's a match of several different styles, choice of words and phrases. Second, LLM has turned a slight plurality into a 100% exclusivity. Say, there are 20 different choices to say the same thing. They are more or less evenly distributed, on…

I understand there is an "Exclude Top Choices" algorithm which helps combat this sort of thing.
Post reply on HN