On AI writing | Hacker News Reader

On AI writing

bolkunov.com

96 pointsdbolkunov62 comments

I love the closing artwork. It's really on theme.

I agree with the author. Content marketers, journalists and other professional writers are very good at making content sound engaging without losing the "voice" of their clients. Tech writers are generally good at presenting complex topics in very accessible ways. They develop style guides that make this easier to do at scale, and editors who are usually better at reading and tweaking over writing from scratch help put everything together.

AI is certainly able to ingest all of that and create similar work, but doing so requires A LOT of hand-holding. Without that, LLMs bias towards writing like a cross between an exec assistant who just started using Grammarly and a junior beat journalist trying to prove something no matter the framing.

Most people don't do that, mostly because writing is a means to an end for many; thus, I've seen the three-banger punchline style in academic works and Claude's EXTREMELY tired "we have Axios at home" headlines in mediums you'd never usually see them in.

What usually does me in is that LLMs take FOREVER to get their "point" across. Whiz-bang gotcha prose only works if you don't spend five paragraphs telling me something that could have gotten distilled into a two-sentence blurb.

What's worse is that summarizing with AI is really popular but doing so usually leaves out important details, and you never know exactly what unless you skim through the piece. When that piece is AI-generated, as they are increasingly becoming...yeah.

Interesting perspectives, a couple of questions that I've been pondering, and am currently curious of your perspectives on

1. It seems we are starting to see more people write like ai, is this something you have measured yet? (I never used em-dash (big fan of the parens), and I resisted for a while, re: ai, but now am trying to use them appropriately)

2. I've been reading from paper again, and my own writing style is evolving, is this too something you can/have measured?

I think someone "writing like ai" is not really the same thing as the flaws we see from ai. "Writing like ai" might mean we adopt certain conventions in how we communicate or frame ideas, but the ai flaws are more along the lines of weasel words or bullshittery; avoiding incisive ideas in favor of fluffiness, or in favor of pithy-sounding phrases that communicate nothing. That's a problem that humans have had long before AI.

Indeed, I'm currently reading Gravitas: Communicate with Confidence, Influence, and Authority by Caroline Goyder. It's phenomenal and I highly recommend it.

> (I never used em-dash (big fan of the parens), and I resisted for a while, re: ai, but now am trying to use them appropriately)

I started using the emdash after learning of the compose key feature of X11. Not tuning the emdash dial to 11 or anything, but occasionally. Now I avoid it to spare myself the indignity of someone confusing me for a chatbot. In general I prefer a more vulgar style of writing as a direct response to those chatbot tics.

Of course, that now distinct LLM style was kind of a thing before, too, but if you weren't writing some bullshit on SEO on Medium or congratulating yourself for your achievements on LinkedIn, using elements of that style wasn't such a source of confusion. Now the chatbots will frame everything as though it's an insight on B2B sales strategies that struck them when they tripped on a dog leash while jogging to work.

your comment was a pleasure to read :]

Some of us old enough to learn to type on keyboards without compose keys (and OSes pre unicode) learnt to use multiple dashes for en- and em-dashes -- like this. Like two spaces after a period, this is ingrained muscle memory that I don't consciously think about when typing.

I'm very glad of this now, because it means that I can still write with en- and em-dashes, while also demonstrating that my writing is not LLM-generated.

Definitely noticed influence of AI on my writing (and probably on my speaking as well). I probably do not like most things I pick up from it unconsciously. However, when I use AI to critique or spellcheck me - I find it's much more useful and that I've learnt something. Though it's not exactly AI writing then.

This is why I always use the 2x single dashes instead of em-dash

haha, I have instructions in my user AGENTS.md to use single dashes instead of em-dash unicode char

> When you know the writer personally, it becomes harder to see the text from the author’s perspective if you know it’s not her style.

I think this is why I absolutely hate reading AI-generated PR descriptions, Readmes or wiki entries at work. I work with these people every day, we talk to each other in meetings, but then I read perfectly formatted Slop in the freshly opened PR that has your name and picture attached to it, yet it's almost impossible for me to care about it because (subconsciously or not) i can't bring myself to see it as yours. Same or even worse for reviews.

For me long texts (internal docs) were definitely the hardest part. However, my colleagues were nice and all agreed to limit AI usage for writing docs that are for other people to read (and not for agents).

We agreed not to use AI-based communication internally, including PR descriptions. I’d rather not have a description at all than an AI-based one.

I don’t understand how people aren’t embarrassed putting those kind of PRs online tbh.

AI assisted writing is so popular despite being inferior, because writing well is inherently hard. Good writing requires research, organization, style/grammar and other aspects like "turn of phrase", and all have to work together. Sit down to write something and it can feel overwhelming looking at the blank page and not knowing what to do.

Same. I hate AI writing. AI code is okay. I can deal with that. But I want people to be able to explain things themselves. It's especially bad when some folks use a prompt to generate an even larger prompt, then pretend they "wrote a spec." Meanwhile, the spec is virtually incomprehensible gibberish about gaps, evidence, and load bearing dependencies.

Some say "give me the prompt" and I will do it myself. That would save time if we just shared prompts + model used.

It undermines an assumption that until recently was nearly impossible to violate: that the author spent time and effort to produce the writing you are spending time and effort to read.

> Personally, I feel that I am at least a third slower reading AI-generated text, even if the length is the same.

I've been reading an AI-generated technical proposition. It was even slower for me.

> Beyond this, most models are extremely verbose, which makes total reading time much higher.

Painfully true. When I took a semster of law studies and got my first homework assignment, I was flabbergasted by how short the submission length limits were. I could barely write an introduction to my mini-paper with that number of pages! And I was expecteds to cover everything, and have an explicit conclusion, and a complete and meticulous bibliography. I struggled sooo much to be more concise and to let go of less-critical points I wanted to make! But it was a great intellectual and rhetorical exercise.

AI always seems to add extra words as if it’s trying to prevent being misunderstood. human writing leaves stuff unsaid because it’s assumed people can infer.

Earlier this year I got asked to take over on leading a project where all the planning artifacts were AI generated.

They were mostly incomprehensible, mutually contradictory, and only tenuously related to the problem we were trying to solve. The poor engineering team had wasted god-only-knows-how-many person-hours building the tool, only for beta users to be completely unable to figure out how to actually use it. Getting things back on track was quite the process. I would estimate that using AI cost the company about 100 times as many person-hours as the original project lead saved themself by avoiding doing any of their own thinking and writing.

It is interesting how despite huge and rapid advances in coding or math, AI still sucks at writing and is (mostly) obvious, at least if people are trying to pay attention. it cannot fool pangram, even though it's not clear exactly what it flags as being AI. Maybe one day AI will close the gap.

I've had some similar thoughts, I find it super fascinating how easy it is for most people to identify AI writing.

In the past two decades of CG etc. untold numbers of words have been spilt on the subject of the Uncanny Valley and you always get this explanation of how the brain has dedicated pattern matching circuitry and can easily flag when faces feel "off".

How interesting that the same is true for language. Turns out we're not just visual-heavy animals, but also very symbolic ones.

i don't see the value in AI closing that gap any more than i see the value in AI closing the gap on talking to my friends for me, or closing the gap on watching a movie for me, or closing the gap on taking a walk on the beach for me, etc, etc, etc; the document is the artifact of having written but the artifact is not the value

I think in coding and math you have a formal system to say what is good and what is bad, while with writing it's much more subjective and thus harder to optimise. Also I believe many of AI advancement in 2026 are linked more to the agents than to models and agents benefit a lot from having environment with clear reward functions

I think Pangram literally just vacuums AI slop into its training data to refine how AI writing is detected. More of it is becoming available by the day, and they have some default eval prompts they used if the methodologies in their paper are still used, so I presume doing this isn't that hard.

AI absolutely has a detectable style once you read enough of it.

I just wish AI was brief.

Don’t add filler words. Be straight to the point.

I think to be brief you need to be precise and it's hard for AI as it's not sure

I think that's part of it. Humans are definitely much more confident (even when wrong) and don't hedge their bets as often.

I think AI also has a lot of internal conversation - you can see it in the reasoning phase - where it will invent shortforms as it talks to itself. This also makes it really hard to follow AI text because it will often invent its own jargon.

LLMs are still just next token predictors. Blasting out words in a forward-only way is how you write a first draft. It's a form of thinking out loud. The ideas are still forming while you generate the text. Generating the text forms the ideas.

Communicating well requires editing. Editing requires going backwards, not just forwards. It requires turning things over in your head. It requires rethinking connections. It requires knowing when one word says more than a sentence. It requires remembering what you were trying to say. It requires remembering.

LLMs don't have much of that. They can mimic some bits with loops, but usually not very well.

It seems like thinking should be the place where it can iteratively draft before creating an output. I've wondered if text diffusion models could in theory write better, thanks to iteratively rewriting/refining the text. It seems to me at least that avenues like that are worth pursuing, so I'm happy to see efforts like Mercury 3 and Diffusion Gemma.

But LLMs don’t think. Neither do diffusion models.

AI always adds extra words. it's annoying how they have not fixed this yet.

The painting at the bottom of the post is actually a really good representation of what is lost with readers and writers both using AI.

I think the signal to noise ratio is high in good writing. Just like in this article and it is very well written. I enjoyed it.

It’s actually really painful to read the ai output, hope it could be better

I've been involved in generating AI fiction / researching AI fiction for a long time now. Both in creating text editors to support it (https://github.com/darrylschaefer/fthr) and in making pipelines to generate it procedurally. I'm also a writer and have practiced both commercial and literary writing, so this isn't something that I've been trying to do purely for money. I have a lot of personal interest in it and have put a lot of time into trying to make interesting things here.

My thoughts on "AI writing" is that the fears around it seem overblown. The frontier models seem to be getting worse, if anything. If one were to make fine-tunes with a specific style, well, that's also a lot of work just to write in a style. How many books in that style can one sell to justify the training? I don't know if the numbers ever add up.

The entire thing might be a great thing for writing and eventually lead to a revival, because now readers will have a motivation to understand the human behind the piece which is always the most interesting part.

We have Ai that can solve any math problem or any technical problem. But we don't have AI that can write naturally like a human. Turns out the purest human intelligence is something no Ai could ever reach. We invented in the past century things like measuring intelligence with iq tests only to realise what kind of bullshit this idea turns out to be in 2026. Turns out we had the wrong mental model of what constitutes intelligence and built machines with super intelligence only in a very narrow sense of the term. We need to radically redefine our idea what human intelligence is.

>We have Ai that can solve any math problem or any technical problem.

Does, "How many p's are in raspberry?" count as a math problem? Because ChatGPT still returns an answer of '2'.

various open weight models now get this right by splitting the word (tokens) with hyphens or newlines

the new version of this is Jev on coin flips and dice rolls

I don't think it's impossible. it's more like the necessary resources have not been invested in making truly human-sounding writing. Writing is much smaller market than coding. Anthropic and Open AI became so valuable from coding agents, not writing ability. Every firm is now using AI for coding. Or too early. But I believe it can be done.

I suspect the difficulty in writing is linked to the difficulty in architecting good, maintainable software. Sure the agents can sling code, but they lack so much when it comes to software development, which also involves a lot of human writing.

Maybe. On the other hand, maybe the reason models aren't good at writing yet is because the labs don't think it's useful or profitable (yet) to focus on training models for writing prose.

I think this is it. early on, no one really know what the biggest potential for AI was, so the focus was on images, short video, and text, as those were the most obvious applications. Then in mid 2025 came the goliath of coding + math. It soon became obvious that AI being more like a smarter version of Mathematica was the future, not "the arts" .

I think it’s because the new post training approach that has enabled a huge improvement on long running tasks and tool / computer use and the underlying latent reasoning applies to verifiable problems that can be solved en mass and repeatedly in isolation.

Generating text humans like, empathising and cold reading the user’s intent, observing something interesting if second hand ally the human experience - none of these can be verified without a human. So they rely on larger pre/train and a larger human curated post train corpus. These are not scalable in the same way.

If the ROI was there, they could do more. Startups like sesame and previously pi showed that a more personable agent than today’s frontier models can be achieved with a sane budget. That’s another unverifiable objective.

But I do think this technology that explains the jagged leaps forward this year does not apply to this problem.

> no Ai could ever reach

Mate we’ve only had AI for like 3 years. “Ever” is the wrong word to use here.

Claude resay that as if I was a mentally challenged child (but without your dumb metaphors) using as few words as possible in your best imitation of a Human speaking casually without jargon and ...

(What it feels like to try and get a decent sentence out of my AI.)

Reply on news.ycombinator.com