Comment by gortok

If I want to participate in a conversation in a language I don't understand I use machine translation. I include a disclaimer that I've used machine translation & hope that gets translated. I also include the input to the machine translator, so that if someone who understands both languages happens to read it they might notice any problems.

When I occasionally use MTL into a language I'm not fluent in, I say so. This makes the reader aware that there may be errors unknown to me that make the writing diverge from my intent.

Non-native English speaker here:

Just use a spell checker and that's it, you don't need LLMs to translate for you if your target is learning the language

Agreed, but if someone uses LLMs to help them write in English, that's very different from the "I asked $AI, and it said" pattern.

> I'm not sure what the solution here

The solution is to use a translator rather than a hallucinatory text generator. Google Translate is exceptionally good at maintaining naturalness when you put a multi-sentence/multi-paragraph block through it -- if you're fluent in another language, try it out!

Maybe they should say "AI used for translation only". And maybe us English speakers who don't care what AI "thinks" should still be tolerant of it for translations.

I have found that prompting "translate my text to English, do not change anything else" works fine.

However, now I prefer to write directly in English and consider whatever grammar/ortographic error I have as part of my writing style. I hate having to rewrite the LLM output to add myself again into the text.

As AIs get good enough, dealing with someone struggling with English will begin to feel like a breath of fresh air.

I think even when this is used they should include "(translated by llm)" for transparency. When you use a intermediate layer there is always bias.

I've written blog articles using HTML and asked llms to change certain html structure and it ALSO tried to change wording.

If a user doesn't speak a language well, they won't know whether their meanings were altered.

one solution that appeals to me (and which i have myself used in online spaces where i don't speak the language) is to write in a language you can speak and let people translate it themselves however they wish

i don't think it is likely to catch on, though, outside of culturally multilingual environments

I wrote about this recently. You need to prompt better if you don't want AI to flatten your original tone into corporate speak:

https://jampauchoa.substack.com/p/writing-with-ai-without-th...

TL;DR: Ask for a line edit, "Line edit this Slack message / HN comment." It goes beyond fixing grammar (because it improves flow) without killing your meaning or adding AI-isms.

To my mind, it's like someone saying "I asked Fred down at the pub and he said...". It's someone stupidly repeating something that's likely stupid anyway.

You can have the same problem with Googling things, LLMs usually form conclusions I align with when I do the independent research. Google isn't anywhere near as good as it was 5 years ago. All the years of crippling their search ranking system and suppressing results has caught up to them to the point most LLMs are Google replacements.

In a work context, for me at least, this class of reply can actually be pretty useful. It indicates somebody already minimally investigated a thing and may have at least some information about it, but they're hedging on certainty by letting me know "the robots say."

It's a huge asterisk to avoid stating something as a fact, but indicates something that could/should be explored further.

(This would be nonsense if they sent me an email or wrote an issue up this way or something, but in an ad-hoc conversation it makes sense to me)

I think this is different than on HN or other message boards, it's not really used by people to hedge here, if they don't actually personally believe something to be the case (or have a question to ask) why are they posting anyway? No value there.

These days, most people who try googling for answers end up reading an article which was generated by AI anyway. At least if you go right to the bot, you know what you're getting.

> When I hear "ChatGPT says..." on some topic at work, I interpret that as "Let me google that for you, only I neither care nor respect you enough to bother confirming that that answer is correct."

I have a less cynical take. These are casual replies, and being forthright about AI usage should be encouraged in such circumstances. It's a cue for you to take it with a grain of salt. By discouraging this you are encouraging the opposite: for people to mask their AI usage and pretend they are experts or did extensive research on their own.

If you wish to dismiss replies that admit AI usage you are free to do so. But you lose that freedom when people start to hide the origins of their information out of peer pressure or shame.

"lets ask the dipshit" is how my colleague phrases it

I disagree. It's not a potential avenue for further investigation. Imo ai should always be consulted

Go ahead, use em—let the haters stew in their own typographically-impoverished purgatory.

Some will blindly dismiss anything using them as AI generated, but realistically the em-dash is only one sign among many. Way more obvious is the actual style of the writing. I use Claude all of the time and I can instantly tell if a blog post I’m reading was written with Claude. It is so distinctive. People use some of the patterns it uses some of the time. But it uses all of them all of the time.

"I asked AI and it said basically the same as you."

>Should asking for sources be banned too?

IMO, HN commenters used to at least police themselves more and provide sources in their comments when making claims. It was what used to separate HN and Reddit for me when it came to response quality.

But yes it is rude to just respond "source?" unless they are making some wild batshit claims.

I actually use LLMs to help me dig up the sources. It's quicker than google and you get them nicely formatted besides.

But: Just because it's easy doesn't mean you're allowed to be lazy. You need to check all the sources, not just the ones that happen to agree with your view. Sometimes the ones that disagree are more interesting! And at least you can have a bit of drama yelling at your screen at how dumb they obviously are. Formulating why they are dumb, now there's the challenge - and the intellectual honesty.

But yeah, using LLMs to help with actually doing the research? Totally a thing.

I'd love to see that for article submissions, as well.

What LLM generate is an amalgamation of human content they have been trained on. I get that you want what actual humans think, but that’s also basically a weighted amalgamation. Real, actual insight, is incredibly rare and I doubt you see much of it on HN (sorry guys; I’ll live with the downvotes).

Same acronym still works, just swap gemini in place of google.

No, linking to chatgpt is not a response. For some sort of questions it (which model exactly is it?) might be better, for some might be worse.

LLM summaries of papers often make overly broad claims [1].

I don't think this is a good example personally.

[1] https://arxiv.org/abs/2504.00025

That's a pretty good example. The summary is actually useful, yet it still annoys me.

But I'm not usually reading the comments to learn, it's just entertainment (=distraction). And similar to images or videos, I find human-created content more entertaining.

One thing to make such posts more palatable could be if the poster added some contribution of their own. In particular, they could state whether the AI summary is accurate according to their understanding.

Yeah, but you get two extremes.

Most often I see these answers under posts like "what's the longest river or earth", or "is Bogota a capital of Venezuela?"

Like. Seriously. It often takes MORE time to post this sort of lazy question than actually look it up. Literally paste their question into $search_engine and get 10 the same answers on the first page.

Actually sometimes telling a person like this "just Google it" is beneficial in two ways: it helps the poster develop/train their own search skills, and it may gently nudge someone else into trying that approach first, too. At the same time slowing the raise of the extremely low effort/quality posts.

But sure, sometimes you get the other kind. Very rarely.

I’ve seen so many SO and other forum posts where the first comment is someone smugly saying “just google it, silly”.

Only that, I’m not the one who posted the original question, I DID google (well DDG) it, and the results led me to someone asking the same question as me, but it only had that one useless reply

Agreed, with a caveat. If someone is asking for an objective answer which could be easily found with a search, and hasn't indicated why they haven't taken that approach, it really comes across as laziness and offloading their work onto other people. Like, "what are the best restaurants in an area" is a good question for human input; "how do you deserialize a JSON payload" should include some explanation for what they've tried, including searches.

[dead]

But if I wanted to ask an AI I would put that into ChatGPT, not ask HN. I would only ask that on HN if I wanted other people's opinions!

You could reply with "Hey you could ask [particular LLM] because it had some good points when I asked it" but I don't care to see LLM output regurgitated on HN ever.

This is like saying "I use a motorized scooter at walmart, not because I can't walk, but because it 'walks' better than I can."

If an LLM writes better than you do, you need to take a long look in the mirror and figure what you can do to fix that, because it's not a good thing.

> if you ban this you would just lose the information that it was generated.

The argument is that the information it generated is just noise, and not valuable to the conversation thread at all.

This is... I'll go with "dystopian". If you're not sure you can properly explain an idea, you should think about it more deeply.

You have to be joking

Meh. Might as well encourage people to post links to search results then too.