Don't post generated/AI-edited comments. HN is for conversation between humans

The dynamics of content production are shifting hard right now. Things that used to signal something interesting are being generated in minutes with little thought. It's getting democratized, but also commoditized.

It's too soon to know how this is going to shake out, so we should resist the temptation to impose rules prematurely. And we should especially not do so out of resistance to change (when has that ever worked out?)

But we'll do what we need to do to keep our heads above water. Example: https://news.ycombinator.com/showlim. I figure pragmatics are fine as long one keeps adjusting.

I feel the same and find myself extending it beyond forums. I've started skipping over articles about AI more and more from authors I normally enjoy reading because so few of those articles end up being particularly interesting or insightful.

AI is obviously an important topic but it has been discussed to absolute death the past couple years and very few people have anything useful to add at this point. Things will of course evolve and change in the near term but someone speculating that maybe this will happen or that will happen isn't very useful.

Given the risks and unknowns I think we should collectively be treating it as a major risk to our economic and national security, and figuring out how to mitigate the downside risks without stifling the upside. But most of the people in power have zero interest in doing that so we're all going to YOLO this in real time.

I've been on HN for 15 years and most of the times 80% of the content is not interesting to me, but i come for the 20%

> Though I do wish we'd see less AI related posts on the front page, they simply aren't sparking curiosity, it is the same wrapped in a different format, a different person commenting on our struggles and wins with AI, the 10th software "rewritten" by an AI.

Exactly. I feel like HN has never been this boring. Enough of the slop, let’s talk about interesting stuff again!

If you haven't yet checked it out, I'd recommend taking a look at Tildes for similarly high quality submissions/conversations as on HN. It really is such a breath of fresh air compared to most other platforms.

I personally joined HN because of various AI discussions.

Comparatively, other sites such as Reddit, Twitter and YouTube just shill content, applications or products. A ton of the posts on Reddit are just AI written ffmpeg wrappers which no one should care about but apparently people do...

This is very well put, and captures my feelings on it. I take it as disrespect that someone would have any expectation for me to read something they can’t be bothered to write. LinkedIn is a great example - my entire professional network is just spamming at this point, which drowns out others that DO put in any effort.

If it takes longer to read, it's not an AI problem, but the author failing to catch that the comment is too drawn out. I don't see how it is a problem to have AI write a comment if you agree with the content. If it is bad content, it will eventually reflect badly on the author anyway.

When I have AI write things for me, I'm spending a good amount of time on it - certainly longer than it takes to read. I'm also usually editing it quite a bit. Maybe I'm an outlier, but I still don't think it's appropriate to make a blanket statement about using AI to write content violating this social contract you described.

Where does the line fall? I can use an LLM to help form new and novel thoughts into prose, right? To structure and present it in conventional language rather than stream of thought. Is that disrespectful? It doesn't feel so.

I guess, in theory, this can eventually be countered by people using LLM browser integrations to tell them whether comments are worth reading (and maybe to summarize long comments). Is anyone currently working on that? It might be interesting to see.

Well just have an AI read it for you then!

That reminds me of the gmail LLM usage where AI can writes your emails for you and also summarize incoming ones. Maybe we lost the thread somewhere...

It's not just about the increase in volume, it's about the delta between the prompt and the generation.

If the generation merely restates the prompt (possibly in prettier, cleaner language), then usually it's the case that the prompt is shorter and more direct, though possibly less "correct" from a formal language perspective. I've seen friends send me LLM-generated stuff and when I asked to see the prompt, the prompts were honestly better. So why bother with the LLM?

But if you're using the LLM to generate information that goes beyond the prompt, then it's likely that you don't know what you're talking about. Because if you really did, you'd probably be comfortable with a brief note and instructions to go look the rest up on one's own. The desire to generate more comes from either laziness or else a desire to inflate one's own appearance. In either case, the LLM generation isn't terribly useful since anyone could get the same result from the prompt (again).

So I think LLMs contribute not just to a drowning out of human conversation but to semantic drift, because they encourage those of us who are less self-assured to lean into things without really understanding them. A danger in any time but certainly one that is more acute at the moment.

[dead]

Thought and creativity won't be averaged away because human beings have a drive for these things. This just raises the bar for it. And why not? We get complacent when not pushed.

Dostoevsky said that if all human knowledge could ever be reduced to 2 + 2 = 4, man would stick out his tongue and insist that 2 + 2 = 5. That was a 19th century formulation—he was a contemporary of Boole. I wonder what the equivalent would be for the LLM era.

But it's a weird kind of average... Not the 3 from 1, 2, 3, 4 & 5 but rather like the bland tv-dinner which tastes non-upsetting for most people.

The soft gaussian blur of all human knowledge.

Perhaps closer to “the mean vector point such that all outbound vectors to different training tests are in sum the smallest”? I assume that’s a property of neural networks anyways, though I’m out of date on current math for them.

If you want a more accurate measure then you should subtract "the sum of all human ignorance" before taking the average.

> I've been feeling more and more that generative AI represents the average of all human knowledge.

No, it's far worse. It's the mode of all human knowledge. The amount of effort you have to put into an LLM to get it to choose an option that isn't the most salient example of anything that could fit as a response is monumental. They skip exact matches for most common matches; it's basically a continuity from when search engines stopped listening to your queries and just decided what query they wanted to respond to - and it suddenly became nearly impossible to search for people who had the same first name as anyone who was famous or in the news.

I've tried a dozen times to get LLMs to find authors for me, or papers, where I describe what I remember about them fairly exactly. They deliver me a bunch of bestsellers and popular things, over and over again, who don't even match at all large numbers of the criteria I've laid out.

It's why they're dumb and can't accomplish anything original. It's structural. They're inherently biased to deliver lowest common denominator work. If you're trying to deliver something original or unusual, what bubbles up is samplings of the slop that surrounds us every day. They're fed everything, meaning everything in proportion to its presence in the world. The vast majority of things are shit, or better said, repetitions of the same shit that isn't productive. The things that are most readily available are already tapped out. The things that are productive are obscure.

You can't even get LLMs to say some words by asking them to "say word X." They just will always find a word that will fill that slot "better." As I said, this is just google saying "did you mean Y?" But it's not asking anymore, it's telling.

edit: It's also why asking it to solve obscure math problems is a dumb test. If the math problem is obscure enough, and there's only one way to possibly solve it, and somebody did it once, somewhere, or referred to the possibility of solving it that way, once, somewhere, you're going to have a single salient example. It's not a greenfield, it's not a white sheet of paper: it's a green field with one yellow flower on it, or a piece of white paper with one black sentence on it, and you're asking it to find the flower or explain the sentence.

edit: https://news.ycombinator.com/item?id=47346901 - I'm late and long-winded.

I feel the same about Claude Code. It's a fast but average developer at just about everything and there are some things that average developers are just consistently bad at and therefore Claude is consistently bad at.

Yes, it’s the "sum" of which you extract an average.

We aren't asking people to not use AI. (We use it ourselves.) What we're asking is not to post AI-generated comments to Hacker News. (We don't do that ourselves.)

By all means make good use of LLMs and other AI. What counts as good use? The world is figuring that out, it will take years, and HN is no exception (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...). We just don't want it to interfere with the human conversation and connection that this site has always been for.

For example, it has always been a bad idea and against HN's rules when users post things that they didn't write themselves, or do bulk copy-pasting into the threads, or write bots to post things.

As I mentioned, the HN mods (who are also the HN devs) use AI extensively and will be doing so a lot more. The limits on that are not technical; they have to do with (1) how much work we still do manually—the classic "no time to do things that would make the things that take all our time take less of it"; and (2) the amount of psychic rewiring that's required—there's a limit to the RoA (rate of astonishment) that any human can absorb. (It's fascinating how technical people are suffering the most from that this time. Less technical people have longer experience being hit by disorienting changes, so for them the current moment is somewhat less skull-cracking.)

Getting this right doesn't mean replacing human-to-human interaction, it means we should have more time for that, and do a better job of supporting HN users generally, as well as YC founders who want to launch on HN, and so on. The goal is to enhance human relatedness, not diminish it.

The mods here have quite a bit of leeway in how they run the site, YC funds it but effectively Dan is lord & master here and I suspect if the mods were to call it quits YC would lose their funnel pretty quickly. There is some balance, fortunately.

But yes, there is some irony there.

Yes a bit ironic, but I am glad they can see that there are times to use AI, and times for human interaction.

No one will ever think that lying that AI output is your own unique creation is a good thing.

> Please don't complain that a submission is inappropriate. If a story is spam or off-topic, flag it.

> If you flag, please don't also comment that you did.

I don't understand why you cut these, they seem important! (I can understand the others, which feel either implied or too specific.)

> I don't think we have to worry about cute animal pictures taking over HN.

Challenge accepted.

I'm curious, just noticed there's no rule requiring comments to be in English, although I've never actually seen any other languages used here. Since the new directive is to write as best you can rather than use AI either to translate or edit, does that imply that one should write either all in another language or in a mix of English and another language? (The latter is especially relevant as many may either only know a technical term in one language, or know the terms in English but not the grammar to connect them.)

edit to add -- I completely agree with you that when one's English is "good enough," it's much better to read the original rather than an LLMs guess at how to polish it. It's just hard to define what that line is, especially for the poster themselves who has no idea what a native speaker can figure out. Would some posts be removed because they are too difficult to make sense of? Or would they be allowed in their native language?

FWIW I think “Please don't complain that a submission is inappropriate. If a story is spam or off-topic, flag it.” is different from the others.

It’s an instruction for how to use the site. It’s helpful to have it in the guidelines for when the flag feature should be used. Without it, the flag link is much more ominous.

Maybe it could be consolidated with the flag-egregious-comments rule?

Edit to add: IMHO it is not at all obvious on this site that flagging stories is meant to be roughly the equivalent of downvoting comments (and that flagging comments doesn’t have a counterpart at the story level).

I’m really curious how this will go. I have a suspicion that we will see more and more accounts all over the internet being controlled by AI agents and no amount of moderation will be able to stop it.

I'd be a wee bit cautious with the "AI edited" part of it; since that might exclude a number of people with disabilities or for whom english is a second (or third, or later) language.

My reading is that the intent is to have a human voice behind the text.

Monitor and see how it goes I guess!

I suppose I should put my comment here instead of at top level.

Exactly when was this point added? It seems somehow not new, but on the other hand it was missing from an archive.today snapshot I found from last July. (I cannot get archive.org to give me anything useful here.)

Edit:

> Please don't complain that a submission is inappropriate. If a story is spam or off-topic, flag it.

> If you flag, please don't also comment that you did.

Perhaps these points (and the thing about trivial annoyances, etc.) should be rolled up into a general "please don't post meta commentary outside of explicit site meta discussion"?

We're going to add that. I've resisted adding reasons-for-flagging for years, but even I can change my mind every decade or so.

A nice side effect is that it will double as a confirmation step, solving the FFF (fat finger flagging) problem.

‘Flag’ is an algorithmic flag only, and there are no humans in the flag algorithm’s processing loop. They may monitor and react to the ‘queue’ of flagged articles, and they can do special mod things with flagged posts. But if you want to report a guidelines violation for AI-assisted writing to the mods, just email the mods (contact link in the footer) subject “AI-assisted writing flag” or similar with a link to the post/comment. It works, I know, I’ve done it before. It takes maybe 60 seconds and there is no other way on the site (seemingly by OG design!) to guarantee human review but that email.

I’ve actually been thinking about this exact idea for https://hcker.news/. Stay tuned, I’ve already started rolling out some comment filtering.

Thanks for putting this so nicely! We'd much rather hear you in your own voice, and the cost of a few mistakes is far less than the cost of losing that.

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

As a non native speaker, I sometimes use LLMs to search for a way to formulate my thoughts like I intend them to be received by the reader. I'd never just copy the verbatim LLM output somewhere, it always sounds blunt and not like me, but I gladly apply grammar corrections or better phrasing.

I'd normally not do this for a text of this length, but just for fun, here's what ChatGPT suggests:

As a non-native speaker, I sometimes use LLMs to help me find wording that conveys my thoughts the way I want them to be understood by the reader. I would never copy the output verbatim, because it often sounds blunt and unlike me, but I’m happy to use grammar corrections or improved phrasing.

If you're referring to speaking in English - in general I think there is a huge amount of flexibility for making mistakes in English. I'm a native speaker, I am so used to hearing various levels of English from different nationalities that i'm almost blind to it. I much prefer to hear someones true voice even if there are a few inaccuracies, so much of a person's personality is conveyed through their quirks and mistakes.

> ... in experiments in which all outer sensation is withdrawn, the subject begins a furious fill-in or completion of senses that is sheer hallucination. So the hotting-up of one sense tends to effect hypnosis, and the cooling of all senses tends to result in hallucination.

Must quote the last paragraph of Chapter 2: "Hot and Cold media", from Marshall McLuhan's Understanding Media, which I've double-underlined.

For it simultaneously explains to me; TikTok (quick consume-scroll-like-react-"create" dopamine hit cycles) and LLMs (outsourcing the essential mechanical friction of thinking (which requires all senses, for me at least))...

The essential friction of deliberate, first-party speech-making---misspellings and all---is why voice and conversation contains life.

Even if you make mistakes, it often can still be understood. 100% I would rather read your own words, even if they're messy, and ask clarifying questions for what I don't understand

Forrest would be so silent if only the best birds would sing.

You write well enough to use your own voice.

I don’t think it is so binary black/white though.

I don’t mind if someone who has no command of English uses a translator. But there is a difference between a translator and an AI/LLM.

This appears to be leading to people being super quiet about their AI usage. It really feels as if everyone is using it massively but keeping quiet about it. This is a guess as I haven't gone around and asked every single person about their AI usage.

I am reminded about a question I posted in a Vintage Apple subreddit. I described the problem and all the steps I took to try and resolve it. In the middle of the text I also hinted that I asked AI and that it gave be a wildly strange answer which I dismissed but that it gave me hints to continue onwards.

The majority of answers were focused around that one sentence and completely ignoring the rest of the post(and even the problem I was posting about). I was ridiculed (sometimes aggressively) for even considering trying the AI. Eventually someone finally answered the question, I thanked them and continued to get downvoted massively.

While I get that the vintage community can attract some colorful characters this was an interesting observation at how badly they reacted to the post. I've since refrained from mentioning AI and furthermore, trying to limit my involvement with communities like that and ironically working on better ways to use AI to solve problems so as to minimize dealing with them(finding ways of providing more system level data to the AI in my prompt).

Totally agree with you. I come here to read comments made by humans. If I need to read comments made by AI Bots I would go to Twitter or reddit, both made me not read the comments section entirely.

Same here, and similarly, I come here to find interesting submissions from smart people. I want to read their own thoughts in their own words, not what an LLM has to say. I'm capable of prompting my own LLM with their prompts if they'd supply them.

It would be great if we could have some kind of indicator that a submission is AI output, perhaps a submitter could vouch that their submission is AI or not, and if they consistently submit AI spam, they have their submission ability suspended or get banned.

Agreed- if it wasn't important enough to spend the time thinking of a satisfying way of writing it, I don't feel like it's important enough for me to spend my bandwidth reading it.

Not to mention, so much of my thinking has been helped by formulating ways of communicating my thoughts that anyone who isn't in the habit of at least struggling with it is, from my point of view, cheating themselves.

great idea, but seems a little futile if there is no protection agains llms training on HN comments. ironically, if HN can succefully prevent llm content, it will become one of the best sources available for training data

I actually do something similar on my personal site using this note that includes a purposeful typo: https://jasoneckert.github.io/site/about-this-site/

I'm hoping people catch that typo after reading "every single word, phrase, and typo (purposeful or not)" and smiled every time I've had someone post a PR with a fix for it (that I subsequently reject ;-)

Yes, I find LLM-written posts valueless because I can already talk to a LLM any time I want (and get the same info). It's not these commenters are the Queen of Sheba bearing a priceless gift of LLM slop. That stuff's pretty cheap.

Copy+pasted LLM output is actually far worse than prompting an LLM myself, because it hides an important detail: the prompt. Maybe the prompter asked their question wrong, or is trolling ("only output wrong answers!"). I don't know how the blob of text they placed on my screen was generated, and have to take them at their word.

I try to "think my own thoughts" but then I see them elsewhere all the time.

My twitter bio has been "Thoughts expressed here are probably those of someone else." for over half a decade.

Amen and agreed 100%

There is no universal cure so every community has to figure it out. I know HN will.

If the community gets lazy with our standards, we drown.

Downvote & flag the AI slop to hell. If we need other mechanisms, let’s figure those out.

I don't really think that good writing and LLM writing looks all that similar. It's not always easy to spot (and maybe HN users aren't always doing a great job at it), but even the best LLM output tends to have an "LLM smell" to it that's hard to avoid.

Like, sure, LLM writing is almost always grammatically correct, spelled correctly, formatted correctly, etc., which tends to be true of good writing. But there's a certain style that it just can't get away from. It's not just the em-dashes, the semi-colons, or the bulleted lists. It's the short, punchy sentences, with few-to-no asides or digressions. Often using idiom, but only in a stale, trite, and homogenized manner. Real humans, are each different -- which lends a certain unpredictability to our writing, even if trying to write to a semi-formal standard, the way "good" writers often do -- but LLMs are all so painfully the same, and the output shows it.

Good writers are often good in recognizably unique ways. To the extent that LLMs produce “good writing,” which I happen to think they mostly do, they tend to overuse specific devices which give their writing a quality that most people are already sick of.

They look similar. In my experience, they do not read similar at all. You have to pay attention and actually try to appreciate what you're reading. Then, if you try and fail, it might not be your fault.

I use dash a lot while people rather usually use and are used to seeing a hyphen. I was called out on a certain app "wtf dude.. the least u can do is nt use ai". Well, the person was using shorthand and textpeak a lot, so it was already getting nauseating for me, so this outburst helped me eject, but not before I politely asked why they thought so and dash was the trigger along with "all da time crct grmr and spelling". Also "hu da hell writes dis long sentences". Guilty as charged.

> Good writers use semicolons and em-dashes

I use semicolons a lot. If this is the nouveau tell du jour for LLMs then I'm in trouble.

Some things to think about:

* A comment should be judged on its merits mostly, and if a comment seems to be substantive, interesting, or ask a thoughtful question, it should be acceptable. I think some LLM comments look superficially relevant, but a moment's thought can make me wonder if a comment actually added anything to the discussion, or did it sound like a rephrasing or generalization of a topic?

* Unfortunately for decent new users, account age is one metric on which to judge here.

* People who post here, should want to engage on a subject when they can, and disengage and be quiet when they can't. There is nothing wrong if you're not an expert on something, and it is not desired by the people here to have you alt-tab to an LLM to plug in extra perspective. We can all do that on our own.

AI can make output seem very average or low effort as well if it sounds like everything else.

> My only caution is that good writers and LLMs look very similar, because LLMs were trained on a corpus of good writers.

People moving to careless writing for authenticity while good writing will be considered AI? funny. We want authentic human thought but can only detect human style.

This reddit thread that came out today is the perfect inversion of the discussion here: https://old.reddit.com/r/ChatGPTPromptGenius/comments/1rr19k...

You were correct not to post the summary. HN tends to expect readers to invest time in reading and understanding long form content and for community to step into discussions and offer context and explanations when necessary. One of the most important context statements on this site has been “in mice”, posted as a two word comment, elevated to top comment on the post. An AI summary will miss that context altogether while busily calculating a cliffsnote no one wants to read (and could often get you flagged and potentially banned, even before today’s guideline update). If a reader wants an AI summary, they have the same tools you do to generate it by their own hand.

If you have domain familiarity with it, have some personal insight to offer a lens through, or care about the topic deeply enough to write a summary yourself, then go ahead! I almost never post about AI given my loathing of generative ML, but I posted a critical summary in a recent “underlying shared structure” post because it was a truly exciting mathematical insight and the paper made that difficult to see for some people.

Please don’t use AI to reduce the distinctiveness of your writing style. Run on sentences are how humans speak to each other. Excess commas are only excess when you consider neurotypicals. I’m learning French and I have already started to fuck up some English spelling because of it. None of that matters in the grand scheme of things. Just add -er suffix checks to your mental proofreading list and move on with being you.

It sounds like you already know how to improve your comments, how about just doing those things.

Before chatbots, people used to link to Google search result pages as a passive-agressive way to say “the information is out there, go find it, I don’t care about you enough to explain it to you”

Pasting a chatGPT response into a comment, and labeling it as such, feels the same to me.

It is more, not less, insulting than trying to pass an AI response off as your own.

> I though of pasting in Perplexity's summary, saying it was from Perplexity but that I had checked and it was a good summary.

> Would that be OK or would that count as an AI written comment?

The rule seems written to answer this directly.

Absolutely nobody cares what Perplexity has to say about the case - summary or otherwise. If you mention what the case is, I can ask claude myself if I’m interested.

Better yet, post a link to an authoritative source on the case (helpful but not required).

At minimum, verify your info via another source. The community deserves that much at least.

An AI-generated summary adds nothing positive and actually detracts from the conversation.

I'd be fine with treating this like snippets from Wikipedia with citations back to the article. This way, people can manually verify the sources if they so choose.

I would still say no, there is something about finding the words for yourself, even if they aren't as elegant as an Ai can make. It's fine, most humans prefer imperfection.

The point is we don't want to read Ai summaries, we can make one ourselves if we want. Personally, with certainty, I don't want to read one from Perplexity on the basis that they do the Ai for Trump Social. (reverse-kyc if you are not aware)

For some inspiration on why this is meaningful: https://www.npr.org/2025/07/18/g-s1177-78041/what-to-do-when...

This is how I would use/expect AI to be used in HN. I would also like this clarified.

Perplexity supports sharing URL to the thread. I think it's quite natural to link AI summaries like that.

I mostly agree, although we've seen big shifts in the culture towards rule-deviating norms over time. Look at the guidelines for ideological battles or throwaway accounts, for example. And, as always:

> Comments should get more thoughtful and substantive, not less, as a topic gets more divisive.

This is only meaningful if enough people read it and agree

This discussion reminds me of the Paradigms of Power featured in Adiamante by L E Modisett; about consensus, power, morality and society. It’s a good read.