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'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 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.

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

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...

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

given the content of the text is of significant importance, it would matter very little the tone it is presented with.

As a russian I'm repulsed by both english and russian slop the same way

> If someone can't figure out what you're saying, and don't care to try,

This puts the onus of being comprehensible to the reader, which isn't fair I think. If you can't get your point across in a way that is comprehensible, maybe don't post.

One potential use case is for individuals who cannot read or write English. They could use automatic translation to read HN and an LLM to translate their comment into English. One possibly would be to forbid such use.

[dead]

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.

In addition to "Internet points" mentioned above - influence operations, both from nation states (e.g. the PRC 50 Cent Party, and probably the dozen most powerful nations in general), and from gray/black-market marketing companies.

Influence is valuable, and HN is a place that people who are aware of it trust highly.

(AI generation of random comments helps build "trustworthy" accounts that can then be activated when a relevant issue comes up)

[1] https://en.wikipedia.org/wiki/50_Cent_Party

One trend I noticed here and, annoyingly, in my co-op, is that people will take a really dense and complex topic that's either currently engaged in deep conversation with multiple people or ripe for it, and then post a link to a Chatgpt conversation with a tag like "I didn't have time to get my thoughts together but here's a Chatgpt overview/some suggested solutions!" For me that's the equivalent of "I googled that for you," aka extremely rude.

Thanks, if I wanted Chatgpt's middle-of-the-bellcurve ass response I would have put the five seconds of effort in myself to type the question into its input field.

Most comments on here are really well-written. I can imagine someone for whom English is a second language (or a first language but aren't as good at writing as they'd like to be) using an LLM to "keep up." Of course, this sometimes works until they decide to post something without those tools.

Reputation farming -> upvote rings -> black market promotion

Same as always: being right about something

On HN, I sometimes used AI to change the tone of my comments - e.g., to add sarcasm or extra-polished corporate-speak for comical effect. OK, now I won't.

Internet points.

I get the sense the point of the HN rule is to preserve unique human expression, regardless of how someone's communication skills are at a given point. Like, I periodically see articles on HN which have stale turns of phrase and signs of poor LLM use (which then becomes distracting while reading) and then the author sometimes mentioning in the HN comments they used an LLM to 'help' with their post based on some list of points they wanted to communicate. Yet when it's relied on too heavily like that it smothers the author's own voice.

If an opinion/idea is being communicated in the voice of another then something unique to that user has been lost. Like if I were to have a germ of an premise and told someone else about it and I found their thoughts clearer and how they expressed it and then copied how they'd expressed it then I think I'd be at least crediting them. Otherwise our own growth with self-editing and clarity will just atrophy and the internet will be a soup of homogenized ways of expressing things.

Your “unclear or jumbled” but authentic comment is always better than “feels like chewing sand”, normalised and calibrated LLM outs

I just wrote a similar comment elsewhere, but I would much rather just read your jumbled or unclear writing than whatever's output from an LLM. At least I know you meant at one point the words that are written. It's not a grammar test in English class or an academic paper; if you use a few fragments or run-ons, it's not a big deal.

There is a tradeoff for sure.

But, even though I think slippery slope arguments should be used very sparingly, there is a good case for one here.

Also, learning how to communicate better, and learning to listen better, is a real value add to this site. Which would get washed out if both writing, and therefore reading, were spoon fed by models, who are also washing away individuality of expression and nuance of views.

> Sometimes I copy-paste a paragraph into ChatGPT or whatever, to ensure my (aging) thoughts are being communicated in a crystal clear manner. I

Same here. And sometimes, I got downvoted and treated as an LLM — in the name of valuing the human.

To me, what matters is the will behind the words. Ideas and words themselves are cheap (this becomes clearer every day in the AI age) — they're almost nothing until they're executed and actually help someone.

> "The Dao can be told, but what is told is not the eternal Dao. The Name can be named, but what is named is not the true Name." — Laozi, Dao De Jing

Like code we write — it's dead text on a screen until it's running. And what we really care about is the running effect — and that is exactly the reason, the will, behind why we write the code in the first place.

>I use AI for the elements I feel are weak or unclear in the transcription. Sometimes I copy-paste a paragraph into ChatGPT or whatever, to ensure my (aging) thoughts are being communicated in a crystal clear manner. I cannot always point out why I think they are unclear or jumbled.

Your point is well taken.[0]

Personally, I take a different approach. I use a 5 minute delay for comments on HN so I can look at the post after I submit it, but before anyone else sees it.

This gives me the opportunity to read over my comment and the comment to which I've replied to make sure my prose is decent, my point is clear and any typos or other inaccuracies can be corrected.

I don't use LLMs as an editor as I've found that I'm probably a better editor than the average internet user, which is what LLMs represent.

Perhaps that's arrogant of me, but I'm much more comfortable standing by what I write when it's me writing and editing.

[0] Please note that this is most certainly not a swipe at you or anyone else who uses LLMs as an editor. I just have a different perspective which pushes me in a different direction.

Do we really need to see your every half-baked thought on here though? It's okay not to post or to set a high bar for yourself.

Frankly, even without AI, most communities get degraded as they become more popular and the stream of comments becomes overwhelming. Like there are over 1000 comments on this story and let's be honest, most of it isn't adding value. A great many of them are repeats of other posts, so the person didn't read other people's comments either.

The solutions seem to boil down to making the karma system more draconian. Like instead of focused more on downvoting garbage and upvoting gems, the slush of "mid" posts has to be dealt with somehow. Not sure if rate-limiting accounts would make a noticeable difference. Ironically, perhaps AI is also a solution to the issue, since obviously it can, for example, know all the other comments and could potentially assign some value score in the overall context.

I probably wouldn't post this here post either but I'm hitting reply because of the topic at hand...

The removal of the quotes around "better" discards an entire layer of meaning.

It also loses the voice that was present in the 'before' version. Typos/misuses and all. More tangibly, an entire layer of meaning was dropped when it removed the quotes around 'better'.

As a non native speaker, seeing how many natives keep making the "then/than" mistake, I'm comfortable looking dumb.

I only use AI on critical communications, to make sure that the meaning of my message is the right one.

Otherwise I'm fine making mistakes and I encourage people to correct me.

This is going to sound nuts, but I've noticed comments lately with multiple misspellings that seem intentional - it's almost like they're trying to signal that they're human, rather than LLM written. I've started to think it makes them even more likely to be LLM written than not.

Unfortunately a lot of other do not understand (in the double sense).

I recently had to tell the same thing to a coworker who ran his text through ChatGPT, changing the meaning subtly (in the wrong direction) and the tone completely. I'd rather read his honest opinion in ESL-grade English than something an LLM "polished".

Others will understand, but won't regard that as worthy. That's a difference.

I make mistakes pretty often thanks to auto complete on my phone and carelessness. I've had threads derail and been attacked by people who freak out over grammar.

Chads never backspace.