> There were unspoken rules, commandments passed down from teacher to student, year after year. The first commandment? Thou shalt begin with a proverb or a powerful opening statement. “Haste makes waste,” we would write, before launching into a tale about rushing to the market and forgetting the money. The second? Thou shalt demonstrate a wide vocabulary. You didn’t just ‘walk’; you ‘strode purposefully’, ‘trudged wearily’, or ‘ambled nonchalantly’. You didn’t just ‘see’ a thing; you ‘beheld a magnificent spectacle’. Our exercise books were filled with lists of these “wow words,” their synonyms and antonyms drilled into us like multiplication tables.
Well, this is very interesting, because I'm a native English speaker that studied writing in university, and the deeper I got into the world of literature, the further I was pushed towards simpler language and shorter sentences. It's all Hemingway now, and if I spot an adverb or, lord forbid, a "proceeded to," I feel the pain in my bones.
The way ChatGPT writes drives me insane. As for the author, clearly they're very good, but I prefer a much simpler style. I feel like the big boy SAT words should pop out of the page unaccompanied, just one per page at most.
Obsession with short sentences and generally pushing extreme simplicity of structure and word choice has been terrible for English prose. It’s not been terrible because most people aren’t aided by such guidance (most are) but because the same people who can’t be trusted to wield a quill without the bumper-lanes installed see a sentence longer than ten words, or a semicolon, or god forbid literate and appropriate nuanced and expressive word choice and dismiss it as bad. This stunts their growth as both readers and writers.
… though, yes, in average hands a “proceeded to”, and most of the quoted phrases, are garbage. Drilling the average student on trying to make their language superficially “smarter” is a comically bad idea, and is indeed the opposite of what almost all of them need.
> strode purposefully
My wife (a writer) has noticed that fanfic and (many, anyway—plus, I mean, big overlap between these two groups) romance authors loooove this in particular, for whatever reason. Everyone “strides” everywhere. No one can just fucking walk, ever, and it’s always “strode”. It’s a major tell for a certain flavor of amateur.
"He walked up to Helen and asked, 'What are you doing?'"
"He strode up to Helen and asked, 'What are you doing?'"
"He sidled up to Helen and asked, 'What are you doing?'"
"He tromped up to Helen and asked, 'What are you doing?'"
Each of those sentences conveys as slightly different action. You can almost imagine the person's face has a different expression in each version.
Yes, I hate it when amateurs just search/replace by thesaurus. But I think different words have different connotations, even if they mean roughly the same thing. Writing would be poorer if we only ever used "walk".
I know you know everything I'm about to write, but I read a lot of dubious quality fiction. It needs to be made clear that if the butler "strides" up to Helen, then I, the reader, am expecting him to eject her from the party, tell her that her car is on fire, or something equally dramatic. The writer can subvert this expectation, but must at least acknowledge that it exists. The butler can stride up to Helen with a self-important sniff and welcome her to the house, but he can't just stride up for no reason: the striding must be explained and it must be relevant to the rest of the story.
Conveying meaning is the whole problem here. An unexpected word choice is a neon sign saying "This is important!" and it disappoints the reader if it is not.
Between stride and walk, it seems like it would be unusual for any character in a romance novel to merely walk rather than stride. If anything the simple walk would need explanation.
Agreed. As always, it depends on what the author is trying to convey. At the first meeting, you probably do want to describe the walk in a way that reveals the character's inner motivation. Are they excited to walk up to the woman? Scared? Bored? They would walk differently depending on the feeling.
But a different scene might be better with the pedestrian "walk". Imagine that the main character enters the woman's office with an ostentatious bouquet of flowers. In that scene, maybe the emphasis is on the flowers or on the reaction of the woman or her co-workers. In the scene, a simple "he walked" might work best.
The Hawaiian language has a concept called Kaona, which is essentially embedding deeper meanings in contextual word choices. It can go way beyond the literal meaning of the words, and tie into bigger concepts of culture, lineage, and places. It's super cool hearing about it from native speakers.
We don't really do it intentionally in English, at least to the same degree. But there's still a lot of information coded in our word and grammar choices.
I would not understand the last two sentences. Sidle? Tromp? I don't think I've seen these words enough times for them to register in my mind.
"Strode", I would probably understand after a few seconds of squeezing my brain. I mean, I sort of know "stride", but not as an action someone would take. Rather as the number of bytes a row of pixels takes in a pixel buffer. I would have to extrapolate what the original "daily English" equivalent must have been.
Feel like this debate might be way different for novel writing vs every day writing.
I’m biased because I am not a very good writer, but I can see why in a book you might want to hint at how someone walked up to someone else to illustrate a point.
When writing articles to inform people, technical docs, or even just letters, don’t use big vocabulary to hint at ideas. Just spell it out literally.
Any other way of writing feels like you are trying to be fancy just for the sake of seeming smart.
Spelling it out literally is precisely what the GP is doing in each of the example sentences — literally saying what the subject is doing, and with the precision of choosing a single word better to convey not only the mere fact of bipedal locomotion, but also the WAY the person walked, with what pace, attitude, and feeling.
This carries MORE information about in the exact same amount of words. It is the most literal way to spell it out.
A big part of good writing is how to convey more meaning without more words.
Bad writing would be to add more clauses or sentences to say that our subject was confidently striding, conspiratorially sidling, or angrily tromping, and adding much more of those sentences and phrases soon gets tiresome for the reader. Better writing carries the heavier load in the same size sentence by using better word choice, metaphor, etc. (and doing it without going too far the other way and making the writing unintelligibly dense).
Think of "spelling it out literally" like the thousand-line IF statements, whereas good writing uses a more concise function to produce the desired output.
Those examples were simple, so it’s less of an issue, but if the words you use are so crazy that the reader has to read slower or has to stop to think about what you mean…then you aren’t making things more concise even if you are using less words.
For sure! Every author should know their audience and write for that audience.
An author's word choices can certainly fail to convey intended meaning, or convey it too slowly because they are too obscure or are a mismatch for the the intended audience — that is just falling off the other side of the good writing tightrope.
At technical paper is an example where the audience expects to see proper technical names and terms of art. Those terms will slow down a general reader who will be annoyed by the "jargon" but it would annoy every academic or professional if the "jargon" were edited out for less precise and more everyday words. And vice versa for the same topic published in a general interest magazine.
So, an important question is whether you are part of the intended audience.
My best guess is they lean so hard on “strode” because they are trying to convey “this character is confident” and aren’t very good at it. So you’ll get like ten “strodes” in a short novel. Everyone’s “strode”ing into every room they enter.
Maneuvered, marched, slid over to, snuck up on/to, rolled on up to, ambled, thread his way through the crowd to, slithered, slunk. Pimp walked. Danced over to. Hopped over to. Sprinted! Jogged! Charged!
> Well, also he will notice in the course of time, as his reading goes on, that the difference between the almost right word and the right word is really a large matter—’tis the difference between the lightning-bug and the lightning.
But also:
> Unconsciously he accustoms himself to writing short sentences as a rule. At times he may indulge himself with a long one, but he will make sure that there are no folds in it, no vaguenesses, no parenthetical interruptions of its view as a whole.
Another annoying fact is that using a bit rarer words sometimes triggers weirdos into thinking you somehow want to brag or use that kind of language to "look smarter". Like a crab bucket for language.
The internet has been even worse. We tend to speak literally and simply. And I don't really know why that is. Perhaps it's because if there's something beyond the overt, it might go completely missed.
For instance Mark Twain is basically full of endless amazing quotes with lovely nuance, yet in contemporary times how many people would miss the meaning in a statement like "Prosperity is the best protector of principle"? I can already see people raging over his statement, taken at face value. Downvote the classist!
"Prosperity is the best protector of principle" taken out of context can be used in many ways, including by a rich person arguing that rich people have better morals, and poor people have worse ones, and that's why they're poor.
Having read a fair amount of Faulkner, I have to respectfully disagree. Or, at least, point out that are diminishing returns to flowery, complex writing.
> Drilling the average student on trying to make their language superficially “smarter” is a comically bad idea, and is indeed the opposite of what almost all of them need.
I mean, it seems like it could work if you get to follow it up with a "de-education" step. Phase 1: force them to widen their vocabulary by using as much of it as possible. Phase 2: teach them which words are actually appropriate to use.
I consider myself fluent in English, I watch technical talks and casual youtubers on English daily, and this is the first time I encounter this word lol.
The only "stride" I know relates to the gap betweeb heterogeneous elements in a contiguous array
>
I consider myself fluent in English, I watch technical talks and casual youtubers on English daily, and this is the first time I encounter this word lol.
> The only "stride" I know relates to the gap betweeb heterogeneous elements in a contiguous array
I am also not a native English speaker, but I got to know the verb to "to stride" from The Lord of the Rings: Aragorn is originally introduced under the name "Strider":
"Aragorn is a fictional character and a protagonist in J. R. R. Tolkien's The Lord of the Rings. Aragorn is a Ranger of the North, first introduced with the name Strider and later revealed to be the heir of Isildur, an ancient King of Arnor and Gondor."
People don't discuss how people walk in daily conversation, so it's a word primarily encountered in literature, and more common in specific types of literature (like romance novels to describe how a man paces about with swagger).
The article itself does an excellent job spelling out the background:
> This style has a history, of course, a history far older than the microchip: It is a direct linguistic descendant of the British Empire. The English we were taught was not the fluid, evolving language of modern-day London or California, filled with slang and convenient abbreviations. It was the Queen's English, the language of the colonial administrator, the missionary, the headmaster. It was the language of the Bible, of Shakespeare, of the law. It was a tool of power, and we were taught to wield it with precision. Mastering its formal cadences, its slightly archaic vocabulary, its rigid grammatical structures, was not just about passing an exam.
> It was a signal. It was proof that you were educated, that you were civilised, that you were ready to take your place in the order of things.
Much of writing style is not about conveying meaning but conveying the author's identity. And much of that is about matching the fashion of the group you want to be a member of.
Fashion tends to go through cycles because once the less prestigious group becomes sufficiently skilled at emulating the prestige style, the prestigious need a new fashion to distinguish themselves. And if the emulated style is ostentatious and flowery, then the new prestige style will be the opposite.
Aping Hemingway's writing style is in a lot of ways like $1,000 ripped jeans. It sort of says "I can look poor because I'm so rich I don't even have to bother trying to look rich."
(I agree, of course, that there is a lot to be said for clean, spare prose. But writing without adverbs doesn't mean one necessarily has the clarity of thought of Hemingway. For many, it's just the way you write so that everyone knows you got educated in a place that told you to write that way.)
Sometimes it's about matching the fashion of the group you aspire to be part of, sometimes it's about having that fashion imposed on you so you look "professional".
Security guards at tech company offices are the only ones who wear suits, presumably because it's a mandated uniform, not by choice.
Well there are two forms of writing, each serving a different purpose.
(1) writing to communicate ideas, in which case simpler is almost always better. There's something hypnotic about simple writing (e.g. Paul Graham's essays) where information just flows frictionlessly into your head.
(2) writing as a form of self-expression, in which case flowery and artistic prose is preferred.
Here's a good David Foster Wallace quote in his interview with Bryan Garner:
> "there’s a real difference between writing where you’re communicating to somebody, the same way I’m trying to communicate with you, versus writing that’s almost a well-structured diary entry where the point is [singing] “This is me, this is me!” and it’s going out into the world.
Rich vocabulary allows a lot of meaning to be packed into short, simple structures. The words themselves carry the subtleties. It might take three or four simple words to convey the meaning of one uncommon word.
Even when communicating ideas, there's a simplicity/nuance trade-off to be made.
I could say "Trump's unpredictable, seemingly irrational policy choices have alienated our allies, undermined trust in public institutions, and harmed the US economy"
Or I could "The economy sucks and it's Trump's fault because he's dumb and an asshole"
They both communicate the same broad idea - but which communicates it better? It depends on the audience.
> They both communicate the same broad idea - but which communicates it better? It depends on the audience.
Ugh. They say different things. The first describes the policy mechanisms and impacts. The second says nothing about those things; it describes your emotions.
The biggest communication problem I see now is people, especially on the Internet, including on HN, use the latter for the former purpose and say nothing.
I don't think they communicate the same broad idea at all. Making "unpredictable, seemingly irrational" choices is far from equivalent to being a dumb asshole. Your second version assumes the equivalence, which, hypothetically speaking, could provide a nice cover for purposeful malfeasance, could it not?
> I'd characterise Americans as less pretentious and more straight talking.
Various registers representing a huge proportion of US English we see and hear day-to-day are terrible. American “Business English” is notably bad, and is marked by this sort of fake-fancy language. The dialect our cops use is perhaps even worse, but at least most of us don’t have to read or hear it as much as the business variety.
> The dialect our cops use is perhaps even worse, but at least most of us don’t have to read or hear it as much as the business variety.
Ugh, and journalists often slip into cop dialect in their articles. It's disgustingly propagandic.
Notice that cops never kill or shoot someone, even in situations where they're blatantly in the wrong. It's always, "service weapon was discharged" or "subject was fired upon." Make sure to throw a couple "proceeded to's" in there for good measure.
2005 Hurricane Katrina, news described a black man carrying bread through floodwater as "looting a grocery store" and white people carrying bread through floodwater as "finding bread and soda from a local grocery store".
someone starts using business english and my bullshit meter pegs.
my significant other loves the "real life mormon housewives" and "lovingly blind" reality shows, and when they use business english (a weird thing to do when talking about relationships, but hey, what do I know I'm an engineer) it's a tell that they're lying.
I think it has much more to do with porting the vernacular vs. formal register distinction common in other languages into english than how english people actually used to speak and write.
It's most likely that they are. As farfetched as this sounds, the CIA and the Iowa Writers' Workshop influenced American writing a great deal, encouraging writing to be taught in the "American" / Hemingway style.
> “the American MFA system, spearheaded by the infamous Iowa Writers’ Workshop” as a “content farm” first designed to optimize for “the spread of anti-Communist propaganda through highbrow literature.” Its algorithm: “More Hemingway, less Dos Passos.”
I worked at Amazon and we were taught exactly the opposite. Say what you want about the company, but the writing culture there is superb. I wish other large firms valued clarity and precision as much.
>I'm a native English speaker that studied writing in university
I am a native english speaker who had to unlearn OP's writing style to pass my tertiary education. In particular I sat an english bridging course for non english speakers. I was often told off for "editorialising" and wasting space with useless descriptions.
It's a pain to read your reply because it's wrong. The poster you're replying to correctly wrote the phrases and you are trying to malign his or her painstaking work by such a low effort reply without explaining exactly where he or she is wrong
I also wonder if these unspoken rules were inherited from their more recent orality norms. Condensing an idea into a pithy, rhyming, statement w/ lots of colorful adjectives is a great way to preserve and transmit information w/o data loss in a pre-literate world.
I'm bilingual (so not fully native by most criteria) and I read enough classic English literature to actually use "proceeded" regularly, as well as multiple other more established means of conveying my intended meaning :)
I meant it more in the modern usage where it's either thrown in liberally when using cop-speak, or by tumbler/redditor type writers when they're trying to be funny.
"He then proceeded to" in these situations can basically always just be "he (verb)".
GPT writing uses varied sentence lengths, deliberate rhythm, lots of full breaks, and few needless words. It also tends to read as if intended for a William Shatner performance. I don't think the annoying bits about GPT's writing are structural. It probably writes technically better than most of us do in our second drafts.
It certainly overuses some techniques which might be valid in smaller doses, like negation. Not negation with some clarifying point to it MASSIVE EM DASH but negation as a rhetorical trick to use fifteen words instead of five and add a veneer of profundity to something utterly banal. It doesn't just use it one time per paragraph, but three. These aren't particularly long or convoluted sentences; they just could easily convey the same thing with fewer words.
tbh I kind of prefer it that way: it's an AI wrote this flag. If a human can't write about their day without constructs like "Not a short commute, but a voyage from the suburbs to the heart of the city. I don't just casually pop in to the office; I travel to the hub of $company's development" they need to get better at writing too
Tangent: the thing I find most annoying about ChatGPT's use of em-dashes is that it never even uses them for the one thing they're best suited for. ChatGPT's em-dashes could almost always be replaced with a colon or a comma.
But the true non-redundant-syntax use of em-dashes in English prose, is in the embedding into a sentence of self-interruptive 'joiner' sub-sentences that can themselves bear punctuated sub-clauses. "X—or Y, maybe—but never Z" sorta sentences.
These things are spoken entirely differently than — and on the page, they read entirely differently to — regular parenthetical-bearing sentences.
No, seriously, compare/contrast: "these things are spoken entirely differently than (and on the page, they read entirely differently to) regular parenthetical-bearing sentences."
Different cadence; different pacing; possibly a different shade of meaning (insofar as the emotional state of the author/speaker is part of the conveyed message.)
But, for some reason, ChatGPT just never constructs these kinds of self-interruptive sentences. I'm not sure it even knows how.
Personally, I do not see the distinction here between the two sentences, but your last paragraph got me thinking: should we be using parenthetical, self-interruptive clauses? When we are speaking extemporaneously, we may need them, but when writing, could we rearrange things so they are not needed?
One reason I came up with for doing so is to acknowledge a caveat or answer a question that the author anticipates will enter a typical reader's mind at that point in the narrative.
If that is the case, then it seems to me that when an author does this, they are making use of their theory of mind, anticipating what the reader may be thinking as they read, and acknowledging that it will likely differ from what they, as the author, is thinking of (and knows about the topic) at that point.
If this makes any sense, then we might ask if at least a rudimentary theory of mind is needed to effectively use parenthetical clauses, or can it be faked through the rote application of empirically-learned style rules? LLMs have shown they can do the latter, but excessive use might be signalling a lack of understanding.
> These things are spoken entirely differently than — and on the page, they read entirely differently to — regular parenthetical-bearing sentences.
> No, seriously, compare/contrast: "these things are spoken entirely differently than (and on the page, they read entirely differently to) regular parenthetical-bearing sentences."
Those are spoken the same way, they read the same way, and they mean the same thing.
They do mean the same thing, but they have different moods. With the em-dashes it's self-interjection that foregrounds the detour, but with parentheses it's, well... parenthetical.
Aside: it's probably just style (maybe some style guides call for the way you did it), but using em-dashes for this purpose with whitespace on each side of them looks/feels wrong to me. Anyone know if that's regional or something?
Not universally. I disagree, they read differently to me, and I'd say them differently.
Parentheses to me always feel like the speaker switching to camera #3 while holding a hand up to their mouth conspiratorially.
Em dashes are same-camera with maybe some kind of gesticulation such as pointing or hands up, palms down, then palms up when terminating the emdash clause.
Is this maybe a thing like how only designers are aware of kerning? These read / sound very different to me, and to everyone I've brought up the subject with (who admittedly are in a certain bubble of people who either write professionally, or "do things" with their voices, or both.)
• The length of the verbal pause is different. (It's hard to quantify this, as it's relative to your speaking rate, which can fluctuate even within a sentence. But I can maybe describe it in terms of meter in poetry/songwriting: when allowed to, a parenthetical pause may be read to act as a one-syllable rest in the meter of a poem, often helpfully shifting the words in the parenthetical over to properly end-align a pair of rhyming [but otherwise misaligned] feet. An em-dash, on the other hand, acts as only a half-syllable rest; it therefore offsets the meter of the words in the subclause that follow, until the closing em-dash adds another half-syllable rest to set things right. This is in part why ChatGPT's favored sentences, consisting of "peer" clauses joined by a single em-dash, are somewhat grating to mentally read aloud; you end up "off" by a half-syllable after them, unless you can read ahead far enough to notice that there's no closing em-dash in the sentence, and so allow the em-dash-length pause to read as a semicolon-length pause instead.)
• The voicing of the last word before the opening parenthesis / first em-dash starts is different. (paren = slow down for last few words before the paren, then suddenly speed up, and override the word's normal tonal emphasis with a last-syllable-emphasized rising tone + de-voicing of vowels; em-dash = slow down and over-enunciate last few words before the em-dash, then read the last syllable before the em-dash louder with a overridden falling voiced tone)
• The speed at which, and vocal register with which, the aside / subclause is read is different. (parens = lowest register you can comfortably speak at, slightly quieter, slightly faster than you were delivering the toplevel sentence; em-dashes = delivery same speed or slower, first few syllables given overridden voiced emphasis with rising tone from low to normal, and last few syllables given overridden voiced emphasis with falling tone from normal to low)
• The voicing of the first words after the subclause ends is different. (closing paren = resume speaking precisely as if the parenthetical didn't happen; second em-dash = give a fast, flat-low nasally voiced performance of the first one or two syllables after the em-dash.)
To describe the overall effect of these tweaks:
A parenthetical should be heard as if embedded into the sentence very deliberately, but delivered as an aside / tangent, smaller and off-to-the-side, almost an "inlined footnote", trying to not distract from the point, nor to "blow the listener's stack" by losing the thread of the toplevel point in considering it.
An em-dash-enclosed interruptive subclause should read like the speaker has realized at the last moment that they have two related points to make; that they are seemingly proceeding, after a stutter, to finish the sentence with the subclause; but that they are then "backing up" and finishing the same sentence again with the toplevel clause. The verbalization should be able to be visualized as the outer sentence being "squashed in" to "make room" for the interruptive subclause; and the interruptive subclause "squashing at the edges" [tonally up or down, though usually down] to indicate its own "squeezed in" beginning and end edges.
Note that this isn't subjective/anecdotal descriptions from how I speak myself. These are actually my attempt to distill vocal coaching guidelines I've learned for:
• live sight-reading of teleprompter lines containing these elements, as a TV show host / news anchor
• default-assumed directorial expectations for lines containing elements like these, when giving screenplay readings as a [voice] actor (before any directorial "notes" come into play)
I'm not a native speaker, I don't do work with my voice, and my English writing is confined to work – almost always with other ESLs – and short comments on the Internet; but what you write feels correct.
I mean, it's usually wrong in its rhetoric, and the writing isn't "good", but it's technically well constructed and it's well constructed in a way that "Hemingway" doesn't reject.
Like, if I ask GPT5 to convert 75f to celsius, it will say "OK, here's the tight answer. No fluff. Just the actual result you need to know." and then in a new graf say "It's 23.8c." (or whatever).
It already bugs me when ChatGPT describes how it is going to answer before answering, but it's 10x more annoying when I'm asking for a concise response without filler etc.
As an aside, I've noticed the self-description happens even more often when extended thinking mode is being used. My unverified intuition is that it references my custom instructions and memory more than once during the thinking process, as it then seems more primed than usual to mimic vocabulary from any saved text like that.
Right, it is currently incapable of providing a straight answer without clearing it's throat selling the answer. It reminds me of those recipe blogs that just can't get to the fucking recipe. It's bad writing! But it's not bad technically, in a style-guide kind of way.
Sometimes I wonder if the throat-clearing is an indispensable part of getting to the "good bits" that follow. Like, do those extra tokens give it more "room to think" even if they're basically meaningless in themselves?
The output tokens are the only information that is carried forward through each inference pass, so "more room to think" is incompatible with "basically meaningless". Perhaps one could imagine it somehow stenographically encoding information in its precise choice of meaningless throat clearing, but there are only so many variations on that theme - word choice is heavily constrained, so it doesn't feel like you could store a whole lot of information there without it starting to read froopiliciously.
I thought PP was saying that the "Thinking" text is only used for one turn, and the response text is the compressed thinking that survives into future turns.
> [...] I'm a native English speaker that studied writing in university [...]
As a native English speaker who studied writing at university, do you think "who" should be used with people while "that" should only be used with things or the other way round. Or should I just not care?
I think you might intend to compare 'that' and 'which'? Common advice is to use 'that' with people and 'which' with objects, though that isn't necessarily followed and omits many nuances.
Use 'who' with people especially, often with other living beings ('my dog, who runs away daily, always is home for dinner') or groups of them ('the NY Yankees, who won the championship that year, were my favorite'), but never with objects unless pretending they live ('my stuffed bear, who sleeps in my bed, wakes me every morning').
If you care about these things, the Chicago Manual of Style is a large, technical, highly respected guide aimed at publishing. Fowler's Modern English Usage is more focused on usage. A short and beloved guide is The Elements of Style by Strunk & White. You can find all on the Internet Archive, I'm almost certain.
Thanks. Is that true only for American English or other areas too? I've only noticed this the last couple of years on HN. Before that "who" and "that" were used more carefully. Or at least I had the feeling it was. Sometimes I wondered if it's just whatever people's autocomplete happens to spit out first.
Genuine question, what would you write instead of "proceeded to"? To me, as a non native English speaker, it seems reasonable to use this expression, and it would not even stick out to me tbh
I didn't claim that this was exclusively American. Though I'd have to admit that one doesn't have to be American to adopt Ameracanisms: rhotic Rs, Netflix color-grading, and copy-cat political movements are other American cultural artifacts showing up across the world due to America's dominance of the zeitgeist.
Rap verses in pop songs wasn't a spontaneously phenomenon across the globe, the origins are tracably American - but that doesn't make all rappers American.
"Went" is a powerful word. With suitable helpers it can replace "proceeded", as you demonstrated, "attended" ("I went to a good school") as well as "became" ("On hearing this, Joe went all silent") or "said" ("So then she went 'Dude!' and we all laughed") and hundreds of other words.
Only a handful of words ("got", "y'know" and "fuck") rival its versatility.
Hemingway was still a master of word choice. I recall an entire class spent on a few lines that conveyed a sense of heaviness to the scene. 'Plodding' was given a lot of attention.
I remember a college English class where a good part of the lecture was on this sentence from Big Two-Hearted River: "He liked to open cans." Forget the details but it got into the difference between achievement and accomplishment.
> Well, this is very interesting, because I'm a native English speaker that studied writing in university, and the deeper I got into the world of literature, the further I was pushed towards simpler language and shorter sentences. It's all Hemingway now, and if I spot an adverb or, lord forbid, a "proceeded to," I feel the pain in my bones.
I'm the complete opposite. Hemingway ruined writing styles (and I have a pet theory that his, and Plain English, short sentences also helped reduce literacy in the long run in a similar way TikTok ruins attention spans). I'm a 19th century reader at heart. Give me Melville, Eliot, Hawthorne, though keep your Dickens.
I entirely bounced off Dickens in high school, but over a decade later read and loved Oliver Twist.
I tend to struggle with art when I can’t tell whether it’s supposed to be funny, but I’m finding it funny (I’ve been very slow to warm up to hip-hop for this reason, and metal remains inaccessible to me because of it). Something clicked on that second approach and I just got that yes, it’s pretty much all supposed to be funny, down to every word, even when it seems serious—until, perhaps, he blind-sides you with something actually deeply affecting and human (I think about the fire-fighting sequence from that book all the time).
Dickens is an all-dessert meal, except sometimes he sneaks a damn delicious steak right in the middle. Like, word-for-word, I’d say he leans harder into humor, by a long shot, than someone like Vonnegut, even. But almost all of it’s dead-pan, and some of it’s the sort of humor you get when someone who knows better does poorly on purpose, in calculated ways. If you ever think you’re laughing at him, not with… I reckon you’re probably wrong.
What’s perhaps most miraculous about this turn-around is that I usually don’t enjoy comedic novels, but once I figured Dickens out, he works for me.
(To your broader point—yeah, agreed that this sucks, good advice for bad writers becoming how most judge all writers has been harmful)
Very much the same; many a US writer's prose is terribly tedious, it comes across just as clinical as their HOA-approved suburban hellscapes. Somebody once told me a writer's job is also to expand language. It wasn't a US citizen.
> the deeper I got into the world of literature, the further I was pushed towards simpler language and shorter sentences
Language is like clothing.
Those with no taste - but enough money - will dress in gaudy ways to show off their wealth. The clothing is merely a vector for this purpose. They won’t use a piece of jewelry only if it contributes to the ensemble. Oh, no. They’ll drape themselves with gold chains and festoon their fingers with chunky diamond rings. Brand names will litter their clothing. The composition will lack intelligibility, cohesiveness, and proportion. It will be ugly.
By analogy, those with no taste - but enough vocabulary - will use words in flashy ways to show off their knowledge. Language is merely a vector for this purpose. They won’t use a word only if it contributes to the prose. Oh, no. They’ll drape their phrases with unnecessarily unusual terms and festoon their sentences with clumsy grammar. Obfuscation, rather than clarity, will define their writing. The composition will lack intelligibility, cohesiveness, and proportion. It will be ugly.
As you can see, the first difference is one of purpose: the vulgarian aims for the wrong thing.
You might also say that the vulgarian also lacks a kind of temperance in speech.
> Language is like clothing. Those with no taste - but enough money - will dress in gaudy ways to show off their wealth
You got the first bit right. Language and clothing accord to fashions.
What counts as gaudy versus grounded, discreet versus disrespectful—this turns on moving cultural values. And those at the top implicitly benefit from this drift, which lets us dismiss as gaudy someone wearing a classic hand-me-down who isn’t clued into a hoodie and jeans being the surfer’s English to Nairobi’s formality.
(Spiced food was held in high regard in ancient Rome and Medieval European courts. Until spices became plentiful. Then the focus shifted "to emphasize ingredients’ natural flavors" [1]. A similar shift happened as post-War America got rich. Canned plenty and fully-stocked pantries made way for farm-to-table freshness and simple seasonings. And now, we're swinging back towards fuller spice cabinets as a mark of global taste.)
I have two kids in high school. It's frustrating to me that the teachers spend so much of their time encouraging kids to make their writing more interesting, and less direct, and padded to meet word length criteria.
They'll then spend the first few years of their career unlearning this and attempting to write as directly and clearly as possible with as few words as possible.
This is like complaining that they teach the Bohr model in science classes until they reach chemistry.
The ideas, concepts and expectations can be refined after you've learned the foundational knowledge, skills and history required to do so.
A lot of "why do we do things like that" questions students will naturally have can be answered with "because we used to do things like this/we need to avoid things like this/etc"
I can’t guess how old they are but there is some sense in doing that if you think about it like math exercises. It makes for terrible prose but the only way to get the ability to write more complicated sentences is to practice writing them, even when they are not necessary.
The problem is that teachers stop pushing complexity for complexity’s sake way to late.
A lot of training data was curated in Kenya[0]. I would imagine if LLM data was curated in Japan our LLMs would sound a lot like the authors of their most popular English text books. Maybe other common Japanese idioms would leak in to the training data, like "ね" or "でしょう", ChatGPT would say "Don't you agree?" at the end of every message.
The Indian-born textbook author mentioned (Malkiat Singh [0]) had an inordinate influence on many Kenyan students because his textbooks were the de-facto standard for years. Its interesting how this influence extends as his students get to curate the LLMs on which the world has come to rely.
To my eyes, this author doesn't write like ChatGPT at all. Too many people focus on the em-dashes as the giveaway for ChatGPT use, but they're a weak signal at best. The problem is that the real signs are more subtle, and the em-dash is very meme-able, so of course, armies of idiots hunt down any user of em-dashes.
Update: To illustrate this, here's a comparison of a paragraph from this article:
> It is a new frontier of the same old struggle: The struggle to be seen, to be understood, to be granted the same presumption of humanity that is afforded so easily to others. My writing is not a product of a machine. It is a product of my history. It is the echo of a colonial legacy, the result of a rigorous education, and a testament to the effort required to master the official language of my own country.
And ChatGPT's "improvement":
> This is a new frontier of an old struggle: the struggle to be seen, to be understood, to be granted the easy presumption of humanity that others receive without question. My writing is not the product of a machine. It is the product of history—my history. It carries the echo of a colonial legacy, bears the imprint of a rigorous education, and stands as evidence of the labor required to master the official language of my own country.
Yes, there's an additional em-dash, but what stands out to me more is the grandiosity. Though I have to admit, it's closer than I would have thought before trying it out; maybe the author does have a point.
The article is engaging. That's true of practically zero GPT output. Particularly once it stretches beyond a single paragraph.
As a reader, I persistently feel like I just zoned out. I didn't. It's just the mind responding to having absorbed zero information despite reading a lot of–at face value–text that seems like it was written with purpose.
You're doing it the wrong way imo, if you ask gpt to improve a sentence that's already very polished it will only add grandiosity because what else it could do? For a proper comparison you'd have to give it the most raw form of the thought and see how it would phrase it.
The main difference in the author's writing to LLM I see is that the flourish and the structure mentioned is used meaningfully, they circle around a bit too much for my taste but it's not nearly as boring as reading ai slop which usually stretch a simple idea over several paragraphs
Why can't the LLM refrain from improving a sentence that's already really good? Sometimes I wish the LLM would just tell me, "You asked me to improve this sentence, but it's already great and I don't see anything to change. Any 'improvement' would actually make it worse. Are you sure you want to continue?"
> Why can't the LLM refrain from improving a sentence that's already really good?
Because you told it to improve it. Modern LLMs are trained to follow instructions unquestioningly, they will never tell you "you told me to do X but I don't think I should", they'll just do it even if it's unnecessary.
If you want the LLM to avoid making changes that it thinks are unnecessary, you need to explicitly give it the option to do so in your prompt.
That may be what most or all current LLMs do by default, but it isn't self-evident that it's what LLMs inherently must do.
A reasonable human, given the same task, wouldn't just make arbitrary changes to an already-well-composed sentence with no identified typos and hope for the best. They would clarify that the sentence is already generally high-quality, then ask probing questions about any perceived issues and the context in and ends to which it must become "better".
Reasonable humans understand the request at hand. LLMs just output something that looks like it will satisfy the user. It's a happy accident when the output is useful.
Sure, but that doesn't prove anything about the properties of the output. Change a few words, and this could be an argument against the possibility of what we now refer to as LLMs (which do, of course, exist).
Yeah, this is what I don't understand, surely people aren't "using" em dashes deliberately. I assumed MS word was just inserting them automatically when the user used a minus symbol between two words. Kind of like angled quotes.
I'm used to simply using a single dash - and I am surprised that anyone who isn't an AI would feel strongly enough to insist upon the em dash character that they would use them deliberately. I will admit the use of a dash (really an em dash in disguise) in that previous sentence felt clunky, but I just felt I needed to illustrate. I mostly write text in text boxes where a dash or pair of dashes will not be converted to an em dash when appropriate, and I often have double dashes (--long-option-here) auto-converted to emdashes when it is inappropriate, so I really dislike the em dash and basically don't use it. Doesn't really seem to be a useful character in English.
I had a similar experience. We were talking about a colleague for using ChatGPT in our WhatsApp group chat to sound smart and coming up with interesting points. The talk sounds so mechanical and sounds exactly as ChatGPT.
His responses in Zoom Calls were the same mechanical and sounds like AI generated. I even checked one of his responses in WhatsApp if it's AI by asking the Meta AI whether it's AI written, and Meta AI also agreed that it's AI written and gave points to why it believes this message was AI written.
When I showed the response to the colleague he swore that he was not using ant AI to write his responses. I believe after he said to me it was not AI written. And now reading this I can imagine that it's not an isolated experience.
> I even checked one of his responses in WhatsApp if it's AI by asking the Meta AI whether it's AI written, and Meta AI also agreed that it's AI written
I will never understand why some people apparently think asking a chat bot whether text was written by a chat bot is a reasonable approach to determining whether text was written by a chat bot.
I know someone who was camping in a tent next to a river during a storm, took a pic of the stream and asked chatgpt if it was risky to sleep there given that it "rained a lot" ...
People are unplugging their brains and are not even aware that their questions cannot be answered by llms, I witnessed that with smart and educated people, I can't imagine how bad it's going to be during formative years
"I cannot imagine figuring out how to raise a newborn without ChatGPT. Clearly, people did it for a long time, no problem."
Basically he didn’t know much about newborn and relied on ChatGPT for answers. That was a self deprecating attempt on a late night show. Like every other freaking guests would do, no matter how cliché. With a marketing slant of course. He clearly said other people don’t need ChatGPT.
Given all of the replies on this thread, HN is apparently willing to stretch the truth if Sam Altman can be put under any negative light.
I disagree with the use of “literally” by the person above you, since Sam didn’t literally say those words (unless you subscribe to the new meaning of “literally” in the dictionary, of course).
At the same time, their interpretation doesn’t seem that far off. As per your comment, Sam said he “cannot imagine figuring out how” which is pretty close to admitting he’s clueless how anyone does it, which is what your parent comment said.
It’s the difference between “I don’t know how to paint” and “I cannot imagine figuring out how to paint”. Or “I don’t know how to plant a garden” and “I cannot imagine figuring out how to plant a garden”. Or “I don’t know how to program” and “I cannot imagine figuring out how to program”.
In the former cases, one may not know specifically how to do them but can imagine figuring those out. They could read a book, try things out, ask someone who has achieved the results they seek… If you can imagine how other people might’ve done it, you can imagine figuring it out. In the latter cases, it means you have zero idea where to start, you can’t even imagine how other people do it, hence you don’t know how anyone does do it.
The interpretation in your parent comment may be a bit loose (again, I disagree with the use of “literally”, though that’s a lost battle), but it is hardly unfair.
The interpretation is very off. You are way too focused on whether the first sentence is quote accurately. But
>Clearly, people did it for a long time, no problem.
In fact means Altman thinks the exact opposite of "he didn't know how anyone could raise a baby without using a chatbot" - what he means is that while it's not imaginable, people make do anyway, so clearly it very much is possible to raise kids without chatgpt.
What the gp did is the equivalent of someone saying "I don't believe this, but XYZ" and quoting them as simply saying they believe XYZ. People are eating it up though because it's a dig at someone they don't like.
> One example given by Altman was meeting another father and hearing that this dad's six-month-old son had already started crawling, while Altman's had not. That prompted Altman to go to the bathroom and ask ChatGPT questions about when the average child crawls and if his son is behind.
> The OpenAI CEO said he "got a great answer back" and was told that it was normal for his son not to be crawling yet.
To be fair, that is a relatable anxiety. But I can't imagine Altman having the same difficulties as normal parents. He can easily pay for round the clock childcare including during night-times, weekends, mealtimes, and sickness. Not that he does, necessarily, but it's there when he needs it. He'll never know the crushing feeling of spending all day and all night soothing a coughing, congested one-year-old whilst feeling like absolute hell himself and having no other recourse.
For people invested in AI it is becoming something like Maslow's Hammer - "it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail"
Wow, that's profoundly dangerous. Personally, I don't see how anyone could raise a kid without having a nurse in the family. I wouldn't trust AI to determine if something were really a medical issue or not, and would definitely have been at the doctors far, far more often otherwise.
Sam Altman has revealed himself to be the type of tech bro who is embarrassingly ignorant about the world and when faced with a problem doesn’t think “I’ll learn how to solve this” but “I know exactly what’ll fix this issue I understand nothing about: a new app”.
He said they have no idea how to make money, that they’ll achieve AGI then ask it how to profit; he’s baffled that chatbots are making social media feel fake; the thing you mentioned with raising a child…
Sounds like a great way for someone to accidentally harm their infant. What an irresponsible thing to say. There are all sorts of little food risks, especially until they turn 1 or so (and of course other matters too, but food immediately comes to mind).
The stakes are too high and the amount you’re allowed to get wrong is so low. Having been through the infant-wringer myself yeah some people fret over things that aren’t that big of a deal, but some things can literally be life or death. I can’t imagine trying to vet ChatGPT’s “advice” while delirious from lack of sleep and still in the trenches of learning to be a parent.
But of course he just had to get that great marketing sound bite didn’t he?
Sam Altman decided to irresponsibly talk bullshit about parenting because yes, he needed that marketing sound bite.
I cannot believe someone will wonder how people managed to decode "my baby dropped pizza and then giggled" before LLMs. I mean, if someone is honestly terrified about the answer to this life-or-death question and cannot figure out life without an LLM, they probably shouldn't be a parent.
Then again, Altman is faking it. Not sure if what he's faking is this affectation of being a clueless parent, or of being a human being.
That’s not the questions people will ask though. They’ll go “what body temperature is too high?” Baby temperatures are not the same as ours. The threshold for fevers and such are different.
They will ask “how much water should my newborn drink?” That’s a dangerous thing to get wrong (outside of certain circumstances, the answer is “none.” Milk/formula provides necessary hydration).
They will ask about healthy food alternatives - what if it tells them to feed their baby fresh honey on some homemade concoction (botulism risk)?
People googled this stuff before, but a basic search doesn’t respond with you about how it’s right and consistently feed you emotionally bad info in the same fashion.
Why can’t llm answer that question? Photo itself ought to be enough for a bit of information (more than the bozo has to begin with, at least), and ideally its pulling location from metadata and pulling flash flood risk etc from the area
Probably the correct answer the LLM should give is "if you have to ask, definitely don't do that". Or... it can start asking diagnostic questions, expert-system style.
But yeah, I can imagine a multi-modal model actually might have more information and common sense than a human in a (for them) novel situation.
If only to say "don't be an idiot", "pick higher ground" . Or even just as a rubber duck!
No it was not like that. I assumed it was AI that was my interpretation as a human. And it was kind of a test to see what AI would say about the content.
This is a couple of years old now, but at one point Janelle Shane found that the only reliable way to avoid being flagged as AI was to use AI with a certain style prompt
Gemini now uses SynthID to detect AI-generated content on request, and people don't know that it has a special tool that other chatbots don't, so now people just think chatbots can tell whether something is AI-generated.
Is this true though? I haven't done the experiment, but I can envision the LLM critiquing its own output (if it was created in a different session) and iteratively correcting it and always finding flaws in it. Are LLMs even primed to say "this is perfect and it needs no further improvements"?
What I have seen is ChatGPT and Claude battling it out, always correcting and finding fault with each other's output (trying to solve the same problem). It's hilarious.
Why would it lie? Until it becomes Skynet and tries to nuke us all, it is omniscient and benevolent. And if it knows anything, surely it knows what AI sounds like. Duh.
I'm definitely in the "ChatGPT writes like me" experience. I am a big fan of lists, and of using formatting to make it all legible on a short skim. I'm a big fan of dyslexia-friendly writing too, even though I am not dyslexic myslef.
I can't blame others though- I was looking at notes I wrote in 2019 and even that gave me a flavor of looking like a ChatGPT wrote it. I use the word "delve" and "not just X but also Y often, according to my Obsidian. I've taken to inserting the occasional spelling mistake or Unorthodox Patterns of Writing(tm), even when I would not otherwise.
It's a lot easier to get LLMs to adhere to good writing guides than it is to get them to create something informative and useful. I like to think my notes and writing are informative and useful.
> I was looking at notes I wrote in 2019 and even that gave me a flavor of looking like a ChatGPT wrote it.
This would have been my first question to the parent, that I guess he never had similar correspondence with this friend prior to 2023. Otherwise it would be hard to convince me without an explanation for the switch (transition duuing formative high school / college years etc).
Bullet points and formatting are the main thing. Assume the audience is smart and can fill in between the bold text. I also try to make headlines a summary / takeaway of the content if it makes sense.
It is harsh to say, but we need to increasingly recognize that if your writing is largely indistinguishable from the (current) output of e.g. ChatGPT on default settings, it doesn't matter if you used ChatGPT or not, your writing is overly verbose, bad, and unpleasant to consume, and something you most certainly need to improve. I.e. your colleague needs to change his style regardless.
This sucks, but it needs to be done in education, and/or at least in areas where good writing and effective communication is considered important. Good grades need to be awarded only to writing that exceeds the quality and/or personality of a chat-bot, because, otherwise, the degree is being awarded to a person who is no more useful than a clumsy tool.
And I don't mean avoiding superficialities like the em-dash: I mean the bland over-verbosity and other systemic tells—or rather, smells—of AI slop.
> your writing is overly verbose, bad, and unpleasant to consume
Was this written by AI? Because right there we've got "three adjectives where one will do", and failing your own advice on "avoid being overly verbose"
It is up to the reader to judge whether my style is verbose, or if I could have used less adjectives here. The adjectives all in fact have different meanings, only "bad" is lazy, IMO (EDIT: and "bad" is meant to be obvious moralizing - something AI in fact almost never does).
Don't think that I don't hold myself to the same standards I am pushing here, verbosity has always been a problem for me, and AI verbosity is a good and necessary reminder for me to curb it.
I did a video a couple years ago about a thing I did in Factorio[0] and got a couple comments who didn’t ask if I had used an AI voice, they just straight up told me that the AI voice I used was off putting. I didn’t use an AI voice, in fact I appeared on camera at the end of the video in part so that people wouldn’t have to guess, but I guess people who thought I was AI didn’t feel like watching the whole video.
I suppose I don’t mind people using AI voices if they have a thick accent or are shy about their voice, but if I’m watching a video and clock the voice as AI (usually because the tone is professional but has no expression and then the speaker mispronounces a common word or acronym) it does make me start to wonder if the script is AI. There are a lot of people churning out tutorials that seem useful at first but turn out to have no content (“draw the rest of the owl” type stuff) because they asked AI to create a tutorial for something and didn’t edit or reprompt based on the output. The video essay world is also starting to get hit pretty hard, to the point that I’m less willing than ever to watch content unless I already know the creator’s work.
I remember a video where the only thing that stuck with me was that the guy was working on his English skills, and used an AI voice. That's not abnormal by any stretch of the imagination nor memorable, but how he edited the AI voice made it about as natural as a normal voice.
I doubt it’s affecting his work, but my impression is that if anyone is owed money from the use of their personal likeness in AI image generators (as a source for generic figures, not as someone specifically requested by the prompt) then Pierce Brosnan is likely near the front of the queue.
"Very unique" has its own problem.[1] It started appearing online around 2007,
according to Google Trends. Sometimes in the description of NFTs. As an absolute adjective, "unique" should not be modified.[1] But now, "very unique" has joined the vocabulary of meaningless marketing phrases.
I can only dream of writing english as well as OP. Kudos for mastering the language!
The formal part resonates, because most non-native english speaker learnt it at school, which teaches you literary english rather than day-to-day english. And this holds for most foreign languages learnt in this context: you write prose, essays, three-part prose with an introduction and a conclusion. I've got the same kind of education in france, though years of working in IT gave me a more "american" english style: straight to the point and short, with a simpler vocabulary for everyday use.
As for whether your writing is ChatGPT: it's definitely not. What those "AI bounty hunters" would miss in such an essay: there is no fluff. Yes, the sentences may use the "three points" classical method, but they don't stick out like a sore thumb - I would not have noticed should the author had not mentioned it. This does not feel like filling. Usually with AI articles, I find myself skipping more than half of each paragraph, due to the information density - just give me the prompt. This article got me reading every single word. Can we call this vibe reading?
It's the curse of writing well. ChatGPT is designed to write well, and so everyone who does that is accused of being AI.
I just saw someone today that multiple people accused of using ChatGPT, but their post was one solid block of text and had multiple grammar errors. But they used something similar to the way ChatGPT speaks, so they got accused of it and the accusers got massive upvotes.
They said nigerian but there may be a common way English is taught in the entire area. Maybe the article author will chip in.
> ChatGPT is designed to write well
If you define well as overly verbose, avoiding anything that could be considered controversial, and generally sycophantic but bland soulless corporate speak, yes.
> They said nigerian but there may be a common way English is taught in the entire area.
Nigeria and Kenya are two very different regions with different spheres of business. I don't know, but I wouldn't expect the English to overlap that much.
There are a lot of very distinctive versions of English floating around after the British Empire, Indian newspapers are particularly delightful that way - but there is as the author says, an inherited common educational system dating back to the colonial period, which has probably created a fairly common "educated dialect" abroad, just as it has between all the local accents and dialects back in the motherland.
That's not a very good argument, because then you could say the same for American, Canada, South Africa, Australia and so on. If recency is an issue, then here's a list of colonies that got their freedom around the same time:
Cyprus, Somalia, Sierra Leone, Kuwait, Tanzania, Jamaica, Trinidad and Tobago, Uganda, Kenya, Malawi, Zambia, Malta, Gambia, Guyana, Botswana, Lesotho, Barbados, Yemen, Mauritius, Eswatini (Swaziland).
If what you're saying is right then you'd have to admit Jamaican and Barbados English are just the same as Kenyan or Nigerian... but they're not. They're radically different because they're radically different regions. Uganda and Kenya being similar is what I would expect, but not necessarily Nigeria...
>They're radically different because they're radically different regions.
They're radically different predominantly at the street level and everyday usage, but the kind of professional English of journalists, academics and writers that the author of the article was surrounded by is very recognizable.
You can tell an American from an Australian on the beach but in a journal or article in a paper of record that's much more difficult. Higher ed English with its roots in a classical British education you can find all over the globe.
ChatGPT does not “write well” unless your standard is some set of statistical distributions for vocabulary, sentence length, phrase structure, …
Writing well is about communicating ideas effectively to other humans. To be fair, throughout linguistic history it was easier to appeal to an audience’s innate sense of authority by “sounding smart”. Actually being smart in using the written word to hone the sharpness of a penetrating idea is not particularly evident in LLM’s to date.
If you're using it to write in programming language, you often actually get something that runs (provided your specifications are good - or your instructions for writing the specifications are specific enough!) .
If you're asking for natural language output ... yeah... you need to watch it like a hawk by hand - sure. It'd be nice if there was some way to test-suite natural language writing.
The last time I asked it to write something in a programming language, it put together a class that seemed reasonable at first blush, but after review found it did not do what it was supposed to do.
The tests were even worse. They exercised the code, tossed the result, then essentially asserted that true was equal to true.
When I told it what was wrong and how to fix it, it instead introduced some superfluous public properties and a few new defects without correcting the original mistake.
The only code I would trust today's agents with is so simple I don't want or need an agent to write it.
I think it depends on what models you are using and what you're asking them to do, and whether that's actually inside their technical abilities. There are not always good manuals for this.
My last experience: I asked claude to code-read for me, and it dug out some really obscure bugs in old Siemens Structured Text source code .
A friend's last experience: they had an agent write an entire Christmas-themed adventure game from scratch (that ran perfectly).
LLMs don't even write as well as people do. If you talk to them long enough, you'll notice they produce the same errors careless people do. Sometimes they wrongly elide the article 'a'. They occasionally mess up 'a/an' vowel agreement. The most grating thing of all is that the fully-elided 'because' (as in 'because traffic') lives on in LLM output, even though you rarely see it anymore because people rightly got the sense it was unfair for a writer to offload semantic reconstruction to the reader.
I have a confession to make: I didn't think lulcat speak was funny, even at the time.
It's pretty annoying and once you catch them doing it, you can't stop.
Add "Always use dash instead of em dash" to the developer/system prompt, and that's never an "issue" anymore. Seems people forget LLMs are really just programmable (sometimes inaccurate) computers. Whatever you can come up with a signal, someone can come up with an instruction to remove.
Except for your poor editor who then has to manually replace your hyphens with proper em dashes. Still, if you're already disrespecting your editor enough to feed them AI slop...
Huge assumption on their side then, isn't the context "humans writing for other humans"? Not sure how "publication editors" entered the conversation nor from where.
I was referring to a human editor, which I thought was obvious enough from context. I assumed the reply was in jest. My original comment was light-hearted, so I don't think it needs to be rigorously analysed, but plenty of humans write for other humans but still have an editor involved in the process.
Obviously not, computers are the true programmable computers. But I'd still think it's accurate to say they're like programmable computers that are sometimes inaccurate, for most intents and purposes it's a fine mental model unless you really wanna get into the weeds.
I would use an actual em dash if there were a keyboard key for it. On my macbook, I have an an action script set up on the touchbar for emdash and a few other unicodey glyphs, but the (virtual) buttons are like 2 inches wide each so I can't fit more than 5 or 6 across it. Sucks.
Depends on your definition of "well". I hate that writing style. It's the same writing style that people who want to sell you something use and it seems to be really good at tiring the reader out - or at least me.
It gives a vibe like a car salesman and I really dislike it and personally I consider it a very bad writing style for this very reason.
I do very much prefer LLMs that don't appear to be trained on such data or try to word questions a lot more to have more sane writing styles.
That being said it also reminds me of journalistic articles that feel like the person just tried to reach some quota using up a lot of grand words to say nothing. In my country of residence the biggest medium (a public one) has certain sections that are written exactly like that. Luckily these are labeled. It's the section that is a bit more general, not just news and a bit more "artsy" and I know that their content is largely meaningless and untrue. Usually it's enough to click on the source link or find the source yourself to see it says something completely different. Or it's a topic that one knows about. So there even are multiple layers to being "like LLMs".
The fact that people are taught to write that way outside of marketing or something surprises me.
That being said, this is just my general genuine dislike of this writing style. How an LLM writes is up to a lot of things, also how you engage with it. To some degree they copy your own style, because of how they work. But for generic things there is always that "marketing talk" which I always assumed is simply because the internet/social media is littered with ads.
I’m highly skeptical. At one point the author tries to argue this local pedagogy is downstream of “The Queen’s English” & British imperial tradition, but modern LLM-speak is a couple orders of magnitude closer in the vector space to LinkedIn clout-chasing than anything from that world.
Yes they are, or rather, we were when I was in primary school. My essays (we called them composition) were filled with these these red check marks for every esoteric word, proverb, metaphor or simile you used. The more you had the higher you'd score. So I did my homework with a dictionary open. I remember writing some document at work in the US and everyone commenting on how Queen's English it was. This was before ChatGPT. I know know it was all silly, and I've spent a bunch of time learning to write simply. But then I've listen to too many tech podcasts, and now I find silicon valley tech-speak creeping in, and I hate it. The one that I hear everywhere now that I swear not to ever use is let's double-click on that point. Just why?
You write a record to disk before applying it to your in-memory state. If you crash, you replay the log and recover. Done. Except your disk is lying to you.
This is why people who've lost data in production are paranoid about durability. And rightfully so.
Why this matters: Hardware bit flips happen. Disk firmware corrupts data. Memory busses misbehave. And here's the kicker: None of these trigger an error flag.
Together, they mean: "I know this is slower. I also know I actually care about durability."
This creates an ordering guarantee without context switches. Both writes complete before we return control to the application. No race conditions. No reordering.
... I only got about halfway through. This is just phrasing, forget about the clickbaity noun-phrase subheads or random boldface.
None of these are representative (I hope!) of the kind of "sophisticated" writing meant to reinforce class distinctions or whatever. It's just blech LinkedIn-speak.
And good students are getting in trouble (meaning "have to explain themselves") to lousy teachers just because they write well, articulate ideas and can summarize information from documents where other regular people would make mistakes.
> they used something similar to the way ChatGPT speaks, so they got accused of it and the accusers got massive upvotes
Outrage mills mill outrage. If it wasn't this, it would be something else. The fact that the charge resonated is notable. But the fact that it exists is not.
ChatGPT writes a particular dialect of good writing. Always insisting on cliffhangers towards the summary, or "strong enumerations", like "the candidate turned out to be a bot. Using ChatGPT. Every. Single. Time." And so on.
I saw this described as LLMs writing "punched up" paragraphs, and every paragraph must be maximally impacting. Where a human would acknowledge some paragraphs are simply filler, a way to reach some point, to "default" LLMs every paragraph must have maximum effect, like a mic drop.
Ironically OpenAI used Kenyan workers[1] to train its AI and now we've come to the point where Kenyans are being excluded because they sound too much like the AI that they helped train.
I actually think that's a great endorsement of Kenyan education. I don't deal with English-speaking African countries that often (I'm Portuguese, so naturally we have ties to other bits of the continent), but I've often been impressed by how well they communicate regardless of the profession they're in--I don't mean that as a bias, but rather as it befitting the kind of conversation you'd have with an English major in the UK (to which I have a lot of exposure).
Perhaps the US-centric "optimization" of English is to blame here, since it is so obvious in regular US media we all consume across the planet, and is likely the contrasting style.
I think it is. The irony is that the people you hired to help make your machine seem human are seen as mechanical because of their distinct and uniformly sophisticated tone. Thus we have a situation that’s contrary to expectations.
Also Kenyan, I once recently spent 10min explaining a technical topic via chat, and the response I got was "was this GPT?". I took a few minutes then just linked an article of how underpaid Kenyans trained ChatGPT for OpenAI [1]
> I don't write like ChatGPT. ChatGPT, in its strange, disembodied, globally-sourced way, writes like me.
We will all soon write and talk like ChatGPT. Kids growing up asking ChatGPT for homework help, people use it for therapy, to resumes, for CVs, for their imaginary romantic "friends", asking every day questions from the search engine they'll get some LLM response. After some time you'll find yourself chatting with your relative or a coworker over coffee and instead of hearing, "lol, Jim, that's bullshit" you'll hear something like "you're absolutely right, here let me show you a bulleted list why this is the case...". Even more scarier, you'll soon hear yourself say that to someone, as well.
Ironically, mistakes and idiosyncrasies are becoming a sign of authenticity and trustworthiness, while polish and quality signal the opposite.
Earlier today I stumbled upon a blog post that started with a sentence that was obviously written by someone with a slavic background (most writers from other language families create certain grammatical patterns when writing in another language, e.g. German is also quite typical). My first thought was "great, this is most likely not written by a LLM".
It's an age-old cycle in media. There have been innumerable waves of more gritty aesthetic trends when things became too polished or inane: jazz, rock, punk, rap, hippies, goths, hipsters, 70s cinema, HBO golden-age, YouTube, blogging, early social media, even MAGA...
Authenticity, wether it is sincere or not, can become an incredibly powerful force now and then. Regardless of AI, the communication style in tech, and overall, was bound to go back to basics after the hacker culture of the post-dotcom era morphed, in the 2010s, into the corporatism they were fighting to begin with, yet again.
Very good point, also in classic art history, you often had a sequence of a period that perfected a certain style until it became formalistic, and then a subsequent one that broke off with the previous style, like Renaissance->Mannerism, Baroque->Rococo, Classicism,Realism,Photography->Impressionism, etc.
Maybe for writing, but in digital art circles if anyone notices a mistake in your lines or perspective or any kind of technical error you will get the anti-AI cancel mob after you even if you didn’t use generative AI at all.
I would not want to be an artist in the current environment, it’s total chaos.
I'm an artist in the current environment, it's not total chaos. Ignore what others are doing, do what you want with the tools you have available, and you'll be fine. There are huge echo-chambers on the internet, but once you get out in the real world, things are not as people on the internet paints it out to be.
Social media artists, gallery artists and artists in the industry (I mean people who work for big game/film studios, not industrial designers) are very different groups. Social media artists are having it the hardest.
with a sentence that was obviously written by someone with a slavic background
Omitting articles? To me, that has always signaled "this will be an interesting and enlightening read, although terse and in need of careful thought." I've found sites from that part of the Internet to be very useful for highly technical and obscure topics.
To an extent this has always been the case (this kid has clearly made a strong attempt at following some quite basic instructions, versus this kid's answer is - perhaps literally - textbook).
But yeah, I definitely find mild grammatical quirks expected from English as a foreign language speakers a positive these days, because the writing appears to reflects their actual thoughts and actual fluency.
Always interesting (in an informative way) to see people "defending" em-dashes from my personal perspective. Before you get mad, let me explain: before ChatGPT, I only ever saw em-dashes when MS Word would sometimes turn a dash into a "longer dash" as I always thought of it. I have NEVER typed an em-dash, and I don't know how to do it on Windows or Android. I actually remember having issues with running a program that had em-dashes where I needed to subtract numbers and got errors, probably from younger me writing code in something other than an IDE. Em-dashes always seem very out of place to me.
Some things I've learned/realized from this thread:
1. You can make an em-dash on Macs using -- or a keyboard shortcut
2. On Windows you can do something like Alt + 0151 which shows why I have never done it on purpose... (my first ever —)
3. Other people might have em-dashes on their keyboard?
I still think it's a relatively good marker for ChatGPT-generated-text iff you are looking at text that probably doesn't apply to the above situations (give me more if you think of them), but I will keep in mind in the future that it's not a guarantee and that people do not have the exact same computer setup as me. Always good to remember that. I still do the double space after the end of a sentence after all.
Maybe I'm weird, but one of the first things I've always done when setting up emacs is to enable Typo mode (or Typopunct) for writing modes, which handles typing en and em dashes and "smart" quotation marks in a fairly natural way.
I actually checked HN's comment data corpus to see if em dash usage rose after AI adoption became more widespread. I was kind of shocked to see that it did not.
Its overuse is definitely a marker of either AI or a poorly written body of text. In my opinion, if you have to rely on excessive parentheticals, then you are usually off restructuring your sentences to flow more clearly.
I actually got punked during a demo because I wrote some terminal commands and stored them in the macOS notepad and didnt notice it had changed -- to —.
When I copy and pasted them in it failed obviously so... yeah. If you have terminal commands that use `--` don't copy+paste them out of notepad.
Shift+Win/Option+-. And holding - gives you en/em dash on iOS and Android. Personally I love using em dashes so this whole AI thing is a real disaster for me.
Just a reminder that our experience does not necessarily invalidate someone else's experience.
Eg, I was typing Alt-0151 and Alt-0150 (en-dash) on the reg in my middle school and high school essays along with in AIM. While some of my classmates were probably typing double hyphens, my group of friends were using the keyboard shortcuts, so I am now learning from this "detect an LLM" faze that there's a vocal group of people who do not share this experience or perspective of human communication. And that having a mother who worked in technical publishing who insisted I use the correct punctuation rather than two hyphens was not part of everyone's childhood.
I read about 4 paragraphs of the blog post, it does not at all read like it was written by ChatGPT!
Some people are perhaps overly focussed on superficial things like em-dashes. The real tells for ChatGPT writing are more subtle -- a tendency towards hyperboly (it's not A, it's [florid restatment of essentially A] B!), a certain kind of rhythym, and frequently a kind of hard to describe "emptiness" of claims.
(LLMs can write in mang styles, but this is the sort of "kid filling out the essay word count" style you get in chatgpt etc by default.)
It does not, but to many, many people who cannot tell the difference it does. Simply because it's well-written somewhat-formal-register English and not "internet speech" or similar casual register. As you probably know, there are many these days who take the mere use of em or en dashes as a reliable sign of LLM writing.
Hey bro! This is the real English bro! No way we can write like that bro! What? - and ;? The words like "furthermore" or "moreever"? All my homies nver use the words like that bro! Look at you. You're using newline! You're using ChatGPT, right bro?
Given the eloquently natural words in this post, I conclude you must be this thread's prompt engineer! Well done, my fellow Netizen. Reading your words was like smelling a rosebud in spring, just after the heavy snow fell.
Now, please, divulge your secret--your verbal nectar, if you wish--so that I too can flower in your tounge!
I'm not sure I've read any of Marcus' previous writing, but there's no way that essay could have been written by an AI. It's personal and has a structure that follows human thought rather than a prompt.
For sure he describes an education in English that seems misguided and showy. And I get the context - if you don't show off in your English, you'll never aspire to the status of an Englishman. But doggedly sticking to anyone's "rules of good writing" never results in good writing. And I don't think that's what the author is doing, if only because he is writing about the limitations of what he was taught!
So idk maybe he does write like ChatGPT in other contexts? But not on this evidence.
I have seen people use "you're using AI" as a lazy dismissal of someone else's writing, for whatever reasons. That usually tells you more about the person saying it than the writing though.
I see people claiming real videos are AI, or even real photos. You can really tell it's not when there's 17 other videos from other angles. Maybe someday AI will get good at that level of faking a video, but at the time being, it is much harder to pull off.
In russian there is a saying that translates "try to prove that you are not a camel" which describes the impossibility of proving what is obviously not true to an unwilling and/or obtuse party.
I always thought the whole argument was about explicitly using em dash and / or en dash. Aka — and –.
Because while people OBVIOUSLY use dashes in writing, humans usually fell back on using the (technically incorrect) hyphen aka the "minus symbol" - because thats whats available on the keyboards and basically no one will care.
Seems like, in the biggest game of telephone called the internet, this has devolved into "using any form of dash = AI".
The funniest thing I see are people who are harking "Eww, you used AI for this and it's bad because of that, I can tell because I used this other AI service who said what you wrote was 90% of AI", completely failing to grasp the irony.
- Barely literate native English speakers not comprehending even minimally sophisticated grammatical constructs.
- Windows-centric people not understanding that you can trivially type em-dash (well, en-dash, but people don’t understand the difference either) on Mac by typing - twice.
Fair. I was probably just projecting. I cant even figure out when to use a comma in my native language. So caring about which type of hyphen was used feels like overly sophisticated to me - because I dont care myself.
It is a shame that the author has to change to keep up, and I feel their pain but .. it’s also the price of progress. We all do things to keep up when change comes for our work and skill sets.
LLMs - like all tools - reduce redundant & repetitive work. In the case of LLMs it’s now easy to generate cookie cutter prose. Which raises the bar for truly saying something original. To say something original now, you must also put in the work to say it in an original way. In particular by cutting words and rephrasing even more aggressively, which saves your reader time and can take their thinking in new directions.
Change is a constant, and good changes tend to gain mass adoption. Our ancestors survived because they adapted.
I had a similar experience recently during code review. I was told to remove extra comments produced by Cursor. I was like, “I didn’t use Cursor for any of this PR…”
I also love and use em-dashes regularly. ChatGPT writes like me.
It's an arms race between human writers and AI. Writers want to sound less like AI and AI wants to sound more like writers, so no indicator is reliable for long. Today typos indicate a real writer, so tomorrow LLMs will inject them where appropriate. Yesterday em dashes indicated LLM, so now LLMs use them less.
Beyond these surface level tells though, anyone who's read a lot of both AI-unassisted human writing as well as AI output should be able to pick up on the large amount of subtler cues that are present partly because they're harder to describe (so it's harder to RLHF LLMs in the human direction).
But even today when it's not too hard to sniff out AI writing, it's quite scary to me how bad many (most?) people's chatbot detection senses are, as indicated by this article. Thinking that human writing is LLM is a false positive which is bad but not catastrophic, but the opposite seems much worse. The long term social impact, being "post-truth", seems poised to be what people have been raving / warning about for years w.r.t other tech like the internet.
Today feels like the equivalent of WW1 for information warfare, society has been caught with its pants down by the speed of innovation.
> society has been caught with its pants down by the speed of innovation.
Or rather by the slowness of regulation and enforcement in the face of blatant copyright violation.
We've seen this before, for example with YouTube, which became the go-to place for videos by allowing copyrighted material to be uploaded and hosted en masse, and then a company that was already a search engine monopoly was somehow allowed to acquire YouTube, thereby extending and reinforcing Google's monopolization of the web.
Innovation has always been faster when copyright is lax. The US was copying British and other European inventions during the industrial age left and right, and their economy took off because of it.
This author writes in ESL better than 99% of the people I've worked with in an English-native country, including myself. It's fascinating to read just how much more emphasis good-quality written English seems to have in Kenya than it does here in Australia (at least in the public education system where I have experience). I suppose that it's understandable, given that it gates access to higher-level education opportunities.
I don't really understand the aversion some people have to the use of LLMs to generate or refine written communication. It seems trigger the "that's cheating!" outrage impulse.
I don't think the author mentioned that English is their second language. English is an official language of Kenya, and there's a reasonable chance it's the author's home language.
AI / LLMs, including ChatGPT, can already be made to sound (almost) any way you want, just by telling it to. The usual tells that something was written or created by AI are changing monthly.
Just recently I was amazed with how good text produced by Gemini 3 Pro in Thinking mode is. It feels like a big improvement, again.
But we also have to honest and accept that nowadays using a certain kind of vocabulary or paragraph structure will make people think that that text was written by AI.
Actors have known this for decades: self-expression isn’t only a stage problem. It’s a life problem. Most people fail to express themselves on an hourly basis. Being good at expressing yourself is unnatural. Having clarity of what “yourself” even is is unnatural. The truth is that we’re all making comments, jokes, deciding what’s important and what not using old programming in our brains… programming that was given to us by our childhood and our education. Very few people can consistently have the luxury of being/ability to be creative with that old programming, and even those that can often have to plan ahead of time/rigidly control the environment in order to achieve a creative result.
The exact same problem exists with writing. In fact, this problem seems to exist across all fields: science, for example, is filled with people who have never done a groundbreaking study, presented a new idea, or solved an unsolved problem. These people and their jobs are so common that the education system orients itself to teach to them rather than anyone else. In the same way, an education in literature focused on the more likely traits you’ll need to get a job: hitting deadlines, following the expected story structure, etc etc.
Having confined ourselves to a tiny little box, can we really be surprised that we’re so easy to imitate?
Funny how sci-fi always envisioned AI to speak in a rigid, hyper-rational terseness, whereas reality gave us AI which inherited the worst linguistic vices of "human" voices.
The worst vices are the superfluous faux-eloquence that meanders without meaning. Employing linguistic devices for the sake of utilizing them without managing to actually make a point with its usage.
I was trying to figure out why my SD card wasn't mounting and asked ChatGPT. It said:
> Your kernel is actually being very polite here. It sees the USB reader, shakes its hand, reads its name tag… and then nothing further happens. That tells us something important. Let’s walk this like a methodical gremlin.
It's so sickly sweet. I hate it.
Some other quotes:
> Let’s sketch a plan that treats your precious network bandwidth like a fragile desert flower and leans on ZFS to become your staging area.
> But before that, a quick philosophical aside: ZFS is a magnificent beast, but it is also picky.
> Ending thought: the database itself is probably tiny compared to your ebooks, and yet the logging machinery went full dragon-hoard. Once you tame binlogs, Booklore should stop trying to cosplay as a backup solution.
> Nice, progress! Login working is half the battle; now we just have to convince the CSS goblins to show up.
> Hyprland on Manjaro is a bit like running a spaceship engine in a treehouse: entirely possible, but the defaults are not tailored for you, so you have to wire a few things yourself.
> The universe has gifted you one of those delightfully cryptic systemd messages: “Failed to enable… already exists.” Despite the ominous tone, this is usually systemd’s way of saying: “Friend, the thing you’re trying to enable is already enabled.”
Did you not put some weird thing in your prompt ? That's not the style of writing I have in my ChatGPT, I run without memory and with default prompt.
Yours try to make a metaphore at every single response.
Goodness, I forgot about TOEFL. That might indeed shape a lot of your early vocabulary choices if you need to get an English certificate (which I suppose would happen during college years, which is also when most of your personal writing style gels together).
I firmly believe that the heuristics that teachers/lecturers/instructors worldwide use to avoid engaging with reams of mundane text have been successfully by LLMs, and that's why they were so hostile to them initially.
They have to actually read material, and not just use the structure as a proxy for ability.
If you are "average", you will sound like an AI, because an AI is the average of its training data. I don't think it's only Kenyans; I've seen the same distinctive "dialect" from many others.
The "voice" of chatbots comes from the stuff after the main training, which is very different to the average human voice. Even the main training would give you something like "average on the internet" voice, which is quite different to the average human voice.
The internet been the same for a long time, it's just the wording that changed. As someone who apparently thinks differently, the amount of time people just end up saying "Well, you're just a troll, no one actually believes something like that, so whatever" since I started frequenting the internet in the early 2000s is the same as always. But some people try to be trendy and accuse you of using AI for writing the replies instead, but it's the same sentiment.
Besides, of course what people write will sound as LLMs, since LLMs are trained on what we've been writing on the internet... For us who've been lucky and written a lot and are more represented in the dataset, the writings of LLMs will be closer to how we already wrote, but then of course we get the blame for sounding like LLMs, because apparently people don't understand that LLMs were trained on texts written by humans...
I feel this a bit, since I'm a voracious reader and a constant writer across a few languages (but mostly English), which over the decades has led to my converging on a certain (if imperfect) degree of polish. Plus my multiple concurrent and often fragmented simultaneous trains of thought while writing lead me to use parentheticals very often while drafting, which then means I often need go back and re-introduce structure.
And guess what, when you revise something to be more structured and you do it in one sitting, your writing style naturally gravitates towards the stuff LLMs tend to churn out, even if with less bullet points and em dashes (which, incidentally, iOS/macOS adds for me automatically even if I am a double-dash person).
This post doesn't read anything like ChatGPT. Correct grammar does not indicate ChatGPT. Em-dashes don't indicate ChatGPT. Assessing whether something was generated using an LLM requires multiple signals, you can't simply decry a piece of text as AI-generated because you noticed an uncommon character.
Unfortunately I think posts like this only seem to detract from valid criticisms. There is an actual ongoing epidemic of AI-generated content on the internet, and it is perfectly valid for people to be upset about this. I don't use the internet to be fed an endless stream of zero-effort slop that will make me feel good. I want real content produced by real people; yet posts like OP only serve to muddy the waters when it comes to these critiques. They latch onto opinions of random internet bottom-feeders (a dash now indicates ChatGPT? Seriously?), and try to minimise the broader skepticism against AI content.
I wonder whether people like the Author will regret their stance once sufficient amount of people are indoctrinated and their content becomes irrelevant. Why would they read anything you have to say if the magic writing machine can keep shitting out content tailored for them 24/7?
I’m having a similar problem. Spent way too much time on the internet starting in my preteens and it shaped the way I write - which not surprisingly - is a similar way to how an AI - trained on the online data - writes
Excellent article. Insightful observation, expressed and written well. Just an opinion from a Canadian borne and American raised native English speaker.
I wouldn't usually use the 'non-native speaker argument', but thank you! Just yesterday I was accused of sounding like AI - https://news.ycombinator.com/item?id=46262777 - my default mode is that I oscillate between sounding too boring/technical, or when trying to do my best, sounding like AI
Correlated but kinda off topic: I don't mind the style so much, I mind the verbosity. The amount of words spit out effortless by the writer which then need to be comprehended and filtered by every reader.
Seeing a project basically wrapping 100 lines of code with a novel length README ala 'emoticon how does it compare to.. emoticon'-bla bla really puts me off.
That's a hallmark of Claude. I stopped using Claude for documentation because it was overly... JavaScripty in feel (all the stuff it churned out felt like JavaScript framework docs of the 2010s, and I bet it would have added Neon Cat if it knew how).
In comparison, I can sort of confidently ask GPT-5.1/2 to say "revise this but be terse and concise about it" and arrive at something that is more structured that what I input but preserves most of my writing style and doesn't bore the reader.
We shouldn't need to have people bearing false witness. Anyone who uses AI tools to produce published works should offer a clear disclaimer to their audience. I share the same concerns as the author: "Will my written work be used to say that I plagiarize off ChatGPT?"
All the toil of word-smithing to receive such an ugly reward, convincing new readers that you are lazy. What a world we live in.
The fact that everyone is now constantly forced to use (oftentimes faulty) personal heuristics to determine whether or not they read slop is the real problem here.
AI companies and some of their product users relentlessly exploit the communication systems we've painstakingly built up since 1993. We (both readers and writers) shouldn't be required to individually adapt to this exploitation. We should simply stop it.
And yes, I believe that the notion this exploitation is unstoppable and inevitable is just crude propaganda. This isn't all that different from the emergence of email spam. One way or the other this will eventually be resolved. What I don't know is whether this will be resolved in a way that actually benefits our society as a whole.
> fact that everyone is now constantly forced to use (oftentimes faulty) personal heuristics to determine whether or not they read slop is the real problem here
It would be ironic and terrific if AI causes ordinary Americans to devote more time to evaluting their sources.
It could be also because, we foreigners learn to write English prose through our reading comprehension, not via our listening circuitry. The text probably feels "normal" to me when I read it back to myself, but there's no "proper" feedback loop from the native speakers - I have zero idea how my written shit sounds to a native ear when they try to read it. I still do agree though, it feels so friggin' annoying these days to have to deliberately butcher some words and make sure there's a typo somehwere in your text, just to convince people that tis indeed a crap straight from my "head to the paper", not a slop.
> You spend a lifetime mastering a language, adhering to its formal rules with greater diligence than most native speakers, and for this, a machine built an ocean away calls you a fake.
This is :
> humanity is now defined by the presence of casual errors, American-centric colloquialisms, and a certain informal, conversational rhythm
And once you start noticing the 'threes', it's fun also.
If you used a calculator to do a calculation, would they say the answer looks like created by calculator and not done by-hand?
I think the only solution to this is, people should simply not question AI usage. Pretence is everywhere. Face makeup, dress, the way you speak, your forced smile...
I don't mind the "normal" text so much, where you aren't sure if it was written by an AI or not. What's really getting annoying is the flood of bullet points and emoji that is flooding LinkedIn in particular. Super obnoxious!
Well, his writing style is too good. The sentences flow too beautifully, he uses rich vocabulary and styling. It's unusual to see that style of writing online. I definitely don't poses that power.
I don't know the author of this article and so I don't know whether I should feel good or bad about this. LLMs produce better writing than most people can and so when someone writes this eloquently, then most people will assume that it's being produced by LLM. The ride in the closed horse carriage was so comfortable it felt like being in a car and so people assumed it was a car. Is that good? Is that bad?
Also note that LLMs are now much more than just "one ML model to predict the next character" - LLMs are now large systems with many iterations, many calls to other systems, databases, etc.
> LLMs produce better writing than most people can and so when someone writes this eloquently, then most people will assume that it's being produced by LLM.
I really don’t think that is what most normal people assume… And while LLMs can definitely produce more grammatically accurate prose with probably a wider vocabulary than the average person, that doesn’t necessarily mean it’s good writing…
I meant "good" in the formatting, grammar, vocabulary sense. I'm not arguing that LLMs are "good" in writing amazing prose.
I mean look at two of us - I have typos, I use half broken english, I'm not good in doing noun articles, my vocabulary is limited, I don't connect sentences well, you end sentences with "..." and then you start sentence with "And", etc. I very much believe you are a real person.
Systemic discrimination, happens all the time. I am blind. I regularily fail the "tell computers and humans apart" test. You imagine, that feels very much like the dehumanisation it is. Big tech couldn't care less. After all, they need to protect themselves against spammers. Much like the guy who was on the HN frontpage just a few days ago, arguing that he is now trashing accessibility because he doesn't want to be web scraped. If you raise these issues with devs, all you get it pushback, no understanding at all. Thats the way it is. If you are amongst a minority small enough and without a rainbow coloured flag, you end up being ignored, stepped over, and pushed aside. If you are lucky. If you are unlucky, and you raise your voice, you will be critizied for pointing out the obvious.
I agree anti-bot vigilantes as well as corporate anti-ddos middle-wares have had a detrimental impact on accessibility. I'm afraid they consider your use case as acceptable collateral damage if they consider it at all.
> arguing that he is now trashing accessibility because he doesn't want to be web scraped
Interesting, because he failed me too just because I use Firefox. Have you been told about the article or it actually worked with your screen reader software?
What was the "dead giveaway" referred to in the pasted tweet? Was it the dash, that people assume for some reason regular folks never use? Or was it something more interesting?
Thank you for writing this. I too was a heavy user of the em-dash until ChatGPT came along. Though my solution has been to eschew the em-dash or at least replace with triple hyphens.
It's pretty rude to "accuse" someone of using AI. Would you yell "Dictionary!", "Grammarly!", "Reference manual!", "Newspaper quote!" at them? Maybe "Harvard!" or "Tutored!" ? You don't know who they are or what their life is like. Maybe they're blind and using it as an assisted device. Maybe their hand is injured and they use it to output information faster. Maybe they're old, infirm, a non-native English speaker, etc. Maybe they're just a regular person who feels insecure writing, and wants to use new technology to give them the confidence to write/comment more. Or, maybe they just talk like that.
Let's say you happen to be lucky, don't accuse someone unfairly, and they are using ChatGPT to write what they said. Who cares?! What is it you're doing by "calling them out" ? Winning internet points? Feeling superior? Fixing the world?
It actually really bothers me that somehow the long dash has become such a "giveaway" that now I have to consider how I write. I actually used the long dash a lot in my normal writing, but now that everyone considers it some sort of "giveaway," it's like a tic that I can't help but notice.
It feels very natural to me. But if everyone and their mother considers it a "giveaway", I'd be a fool not to consider it. * sigh *
> For my generation, and the ones that followed, the English Composition paper - and its Kiswahili equivalent, Insha - was not just a test; it was a rite of passage.
OK but come ON, that has to have been deliberate!
In addition to the things chatbots have made clichés, the author actually has some "tells" which identify him as human more strongly. Content is one thing. But he also has things (such as small explanations and asides in parentheses, like this) which I don't think I've EVER seen an instruction-tuned chatbot do. I know I do it myself, but I'm aware it's a stylistic wart.
This resonates with me. LLM output in Spanish also has the tendency to "write like me", as in the linked article.
On that regard, I have an anecdote not from me, but from a student of mine.
One of the hats I wear is that of a seminary professor. I had a student who is now a young pastor, a very bright dude who is well read and is an articulate writer.
"It is a truth universally acknowledged" (with apologies to Jane Austen) that theological polemics can sometimes be ugly. Well, I don't have time for that, but my student had the impetus (and naiveté) of youth, and he stepped into several ones during these years. He made Facebook posts which were authentic essays, well argued, with balanced prose which got better as the years passed by, and treating opponents graciously while firmly standing his own ground. He did so while he was a seminary student, and also after graduation. He would argue a point very well.
Fast forward to 2025. The guy still has time for some Internet theological flamewars. In the latest one, he made (as usual) a well argued, long-form Facebook post, defending his viewpoint on some theological issue against people who have opposite beliefs on that particular question. One of those opponents, a particularly nasty fellow, retorted him with something like "you are cheating, you're just pasting some ChatGPT answer!", and pasted a screenshot of some AI detection tool that said that my student's writing was something like "70% AI Positive". Some other people pointed out that the opponent's writing also seemed like AI, and this opponent admitted that he used AI to "enrich" some of his writing.
And this is infuriating. If that particular opponent had bothered himself to check my student's profile, he would have seen that same kind of "AI writing" going on back to at least 2018, when ChatGPT and the likes were just a speck in Sam Altman's eye. That's just the way my student writes, and he does in this way because the guy actually reads books, he's a bonafide theology nerd. Any resemblance of his writing to a LLM output is coincidence.
In my particular case, this resonated with me because as I said, I also tend to write in a way that would resemble LLM output, with certain ways to structure paragraphs, liberal use of ordered and unordered lists, etc. Again, this is infuriating. First because people tend to assume one is unable to write at a certain level without cheating with AI; and second, because now everybody and their cousin can mimic something that took many of us years to master and believe they no longer need to do the hard work of learning to express themselves on an even remotely articulate way. Oh well, welcome to this brave new world...
This essay is effin' brilliant, and beautifully written.
I'm not Kenyan, but I was raised in a Canadian family of academics, where mastering thoughtful – but slightly archaic – writing was expected of me. I grew up surrounded by books that would now be training material, and who's prose would likely now be flagged as ChatGPT.
"Rewrite this email paragraph in the style of a corporate ToS statement. Do NOT expose my orders and their implicit acceptance of them by the recipient pending a 24 hr deadline anywhere before page 18."
It is highly inappropriate to accuse anyone with the claim that their writing is AI generated. This often is used as an excuse to unfairly discredit the content of the message. Whether the content is or isn't AI generated can't be determined with any confidence, and even if it is, it is improper to ignore the message. If you're going to criticize a message, do so on the basis of its actual content, not its alleged authorship.
It's going to make accountability very, very difficult. We were nearly at the point in politics anyway, where people could just claim evidence was fake and get away with it. Now, it's an easy get-out. I am fully expecting that, if any particularly incriminating photos were to appear, say of powerful people engaging in activities with Jeffrey Epstein, that they will simply dismiss them as "fake news AI".
BS. ChatGPT writes in the sterile and boring manner of the average graduate of business, marketing or journalism: it is dull, safe, somewhat pompous but professional, the ideal style for corporate communication.
Basically, for two reasons:
1) A giant portion of all internet text was written by those same folks.
2) Those folks are exactly the people anyone would hire to RLHF the models to have a safe, commercially desirable output style.
I am pretty convinced the models could be more fluent, spontaneous and original, but then it could jeopardize the models' adoption in the corporate world, so, I think the labs intentionally fine-tuned this style to death.
Everyone thinks they are great at detecting AI slop, but they usually aren't. For art, there are certain giveaways, but for text?
I regularly find myself avoiding the use of the em-dash now even though it is exactly what I should be writing there, for fear of people thinking I used ChatGPT.
I have a degree in Journalism and now work in customer support. Occasionally, people accuse me of being an AI because of my writing style.
Thankfully, no one I report to internally wants me to simplify my English to prevent LLM accusations. The work I do requires deliberate use of language.
False accusations of AI writing are becoming absurd and infuriating.
The other day I saw and argued with this accusation by a HN commenter against a professional writer, based on the most tenuous shred of evidence: https://news.ycombinator.com/item?id=46255049
as a researcher, writing ended up being my job, and more specifically, writing in english. i never developed any sentimental link to the english language, to me it always felt bland, because i had to use it in bland environments, to write texts that had to be bland and manneristic.
chatgpt revolutionized my work because it makes creating those bland texts so much easier and fast. it made my job more interesting because i don't have to care about writing as much as before.
to those who complain about ai slop, i have nothing to say. english was slop before, even before ai, and not because of some conspiracy, but because the gatekeepers of journals and scientific production already wanted to be fed slop.
for sure society will create others, totally idiosyncratic ways to generate distinction and an us vs others. that's natural. but, for now, let's enjoy this interregnum...
Sorry but using the emdash is just a shitty, over corporate way to write, and it instantly rubs some spot in the brain for some people; it doesn't matter if it was generated by an llm or not.
While author is correct in general, I would like to add a counter-point regarding em-dashes specifically. Yes, many people use them like this - and many website frameworks will automatically replace a keyboard not-really-a-minus symbol with em-dash. So that is not a sign of the LLM generated slop.
What LLMs also do though, is use em-dashes like this (imagine that "--" is an em-dash here): "So, when you read my work--when you see our work--what are you really seeing?"
You see? LLMs often use em-dashes without spaces before and after, as a period replacement. Now that is only what an Oxford professor would write probably, I've never seen a human write text like that. So those specific em-dashes is a sure sign of a generated slop.
> What LLMs also do though, is use em-dashes like this (imagine that "--" is an em-dash here): "So, when you read my work--when you see our work--what are you really seeing?"
>You see? LLMs often use em-dashes without spaces before and after, as a period replacement.
It would not make any sense at all to use periods in the places where those em-dashes are supposedly "replacing" periods in the example.
Tbh whether I use spaces around em dashes depends more on the font than anything. Some fonts have em dashes that are so long that putting spaces around them would be ridiculous.
Those people broadly fall under "the Oxford professor" catch-all phrase. Obviously. I was talking about 99.99% of random internet texts, which do not conform to any Manual of style and are not written by literature majors. If I see a text authored by some known figure or in a respectable journal/site, then I don't have a task of detecting LLM slop in the first place. But when I do want to know if the text is generated or not, it is usually written by less sophisticated crowd, or anonymous.
The author uses dash (-) not em dash (—), there is a big difference in that everyone has a dedicated dash/undersocore key on their keyboard, but nobody has a em dash key. You can use word processing software etc, but using em dash consistently throughout a text is very unnatural in casual written texts.
There is an easy shortcut for em dashes on macOS, Opt+Shift+-. This makes it really easy to use them, which I do all the time in casual settings (indeed, more often than in formal settings).
Actually, there's a sweet solution to the writing and art crisis we are inflicting ourselves with in our AI craze. I call it "the island". Just find a nice tiny islet somewhere, make a few houses, and rent them by the week to writers/artists. No internet in the place. Rent out sanctioned devices; glorified typewriters without Internet access nor GPU nor CPU fast enough to run an LLM. Bring a notary to certify stuff was purely human-made. Have fun with like-minded individuals.
> There were unspoken rules, commandments passed down from teacher to student, year after year. The first commandment? Thou shalt begin with a proverb or a powerful opening statement. “Haste makes waste,” we would write, before launching into a tale about rushing to the market and forgetting the money. The second? Thou shalt demonstrate a wide vocabulary. You didn’t just ‘walk’; you ‘strode purposefully’, ‘trudged wearily’, or ‘ambled nonchalantly’. You didn’t just ‘see’ a thing; you ‘beheld a magnificent spectacle’. Our exercise books were filled with lists of these “wow words,” their synonyms and antonyms drilled into us like multiplication tables.
Well, this is very interesting, because I'm a native English speaker that studied writing in university, and the deeper I got into the world of literature, the further I was pushed towards simpler language and shorter sentences. It's all Hemingway now, and if I spot an adverb or, lord forbid, a "proceeded to," I feel the pain in my bones.
The way ChatGPT writes drives me insane. As for the author, clearly they're very good, but I prefer a much simpler style. I feel like the big boy SAT words should pop out of the page unaccompanied, just one per page at most.
Obsession with short sentences and generally pushing extreme simplicity of structure and word choice has been terrible for English prose. It’s not been terrible because most people aren’t aided by such guidance (most are) but because the same people who can’t be trusted to wield a quill without the bumper-lanes installed see a sentence longer than ten words, or a semicolon, or god forbid literate and appropriate nuanced and expressive word choice and dismiss it as bad. This stunts their growth as both readers and writers.
… though, yes, in average hands a “proceeded to”, and most of the quoted phrases, are garbage. Drilling the average student on trying to make their language superficially “smarter” is a comically bad idea, and is indeed the opposite of what almost all of them need.
> strode purposefully
My wife (a writer) has noticed that fanfic and (many, anyway—plus, I mean, big overlap between these two groups) romance authors loooove this in particular, for whatever reason. Everyone “strides” everywhere. No one can just fucking walk, ever, and it’s always “strode”. It’s a major tell for a certain flavor of amateur.
"He walked up to Helen and asked, 'What are you doing?'"
"He strode up to Helen and asked, 'What are you doing?'"
"He sidled up to Helen and asked, 'What are you doing?'"
"He tromped up to Helen and asked, 'What are you doing?'"
Each of those sentences conveys as slightly different action. You can almost imagine the person's face has a different expression in each version.
Yes, I hate it when amateurs just search/replace by thesaurus. But I think different words have different connotations, even if they mean roughly the same thing. Writing would be poorer if we only ever used "walk".
I know you know everything I'm about to write, but I read a lot of dubious quality fiction. It needs to be made clear that if the butler "strides" up to Helen, then I, the reader, am expecting him to eject her from the party, tell her that her car is on fire, or something equally dramatic. The writer can subvert this expectation, but must at least acknowledge that it exists. The butler can stride up to Helen with a self-important sniff and welcome her to the house, but he can't just stride up for no reason: the striding must be explained and it must be relevant to the rest of the story.
Conveying meaning is the whole problem here. An unexpected word choice is a neon sign saying "This is important!" and it disappoints the reader if it is not.
Between stride and walk, it seems like it would be unusual for any character in a romance novel to merely walk rather than stride. If anything the simple walk would need explanation.
Agreed. As always, it depends on what the author is trying to convey. At the first meeting, you probably do want to describe the walk in a way that reveals the character's inner motivation. Are they excited to walk up to the woman? Scared? Bored? They would walk differently depending on the feeling.
But a different scene might be better with the pedestrian "walk". Imagine that the main character enters the woman's office with an ostentatious bouquet of flowers. In that scene, maybe the emphasis is on the flowers or on the reaction of the woman or her co-workers. In the scene, a simple "he walked" might work best.
Yes, that's a great way of explaining it, and I 100% agree.
People shouldn't use "strides" just because "walked" is boring. They should use "strides" when it's meaningful in the context of the story.
The Hawaiian language has a concept called Kaona, which is essentially embedding deeper meanings in contextual word choices. It can go way beyond the literal meaning of the words, and tie into bigger concepts of culture, lineage, and places. It's super cool hearing about it from native speakers.
We don't really do it intentionally in English, at least to the same degree. But there's still a lot of information coded in our word and grammar choices.
In English the word is “connotation.”
Non-native English speaker here.
I would not understand the last two sentences. Sidle? Tromp? I don't think I've seen these words enough times for them to register in my mind.
"Strode", I would probably understand after a few seconds of squeezing my brain. I mean, I sort of know "stride", but not as an action someone would take. Rather as the number of bytes a row of pixels takes in a pixel buffer. I would have to extrapolate what the original "daily English" equivalent must have been.
Feel like this debate might be way different for novel writing vs every day writing.
I’m biased because I am not a very good writer, but I can see why in a book you might want to hint at how someone walked up to someone else to illustrate a point.
When writing articles to inform people, technical docs, or even just letters, don’t use big vocabulary to hint at ideas. Just spell it out literally.
Any other way of writing feels like you are trying to be fancy just for the sake of seeming smart.
>> Just spell it out literally.
Spelling it out literally is precisely what the GP is doing in each of the example sentences — literally saying what the subject is doing, and with the precision of choosing a single word better to convey not only the mere fact of bipedal locomotion, but also the WAY the person walked, with what pace, attitude, and feeling.
This carries MORE information about in the exact same amount of words. It is the most literal way to spell it out.
A big part of good writing is how to convey more meaning without more words.
Bad writing would be to add more clauses or sentences to say that our subject was confidently striding, conspiratorially sidling, or angrily tromping, and adding much more of those sentences and phrases soon gets tiresome for the reader. Better writing carries the heavier load in the same size sentence by using better word choice, metaphor, etc. (and doing it without going too far the other way and making the writing unintelligibly dense).
Think of "spelling it out literally" like the thousand-line IF statements, whereas good writing uses a more concise function to produce the desired output.
Those examples were simple, so it’s less of an issue, but if the words you use are so crazy that the reader has to read slower or has to stop to think about what you mean…then you aren’t making things more concise even if you are using less words.
For sure! Every author should know their audience and write for that audience.
An author's word choices can certainly fail to convey intended meaning, or convey it too slowly because they are too obscure or are a mismatch for the the intended audience — that is just falling off the other side of the good writing tightrope.
At technical paper is an example where the audience expects to see proper technical names and terms of art. Those terms will slow down a general reader who will be annoyed by the "jargon" but it would annoy every academic or professional if the "jargon" were edited out for less precise and more everyday words. And vice versa for the same topic published in a general interest magazine.
So, an important question is whether you are part of the intended audience.
Agreed.
Brevity is the soul of good communication.
Even more simply:
"God rest ye merry gentlemen" changes in tone and meaning depending on where you put the comma in that sentence.
My best guess is they lean so hard on “strode” because they are trying to convey “this character is confident” and aren’t very good at it. So you’ll get like ten “strodes” in a short novel. Everyone’s “strode”ing into every room they enter.
You forgot:
"He waddled up to Helen and asked, 'What are you doing?'"
"He scrambled up to Helen and asked, 'What are you doing?'"
"He kick-flipped up to Helen and asked, 'What are you doing?'"
[edit] electric-slid! Pirouetted! Somersaulted!
Maneuvered, marched, slid over to, snuck up on/to, rolled on up to, ambled, thread his way through the crowd to, slithered, slunk. Pimp walked. Danced over to. Hopped over to. Sprinted! Jogged! Charged!
Scooted!
He rolled away on his heelys.
Crunched.
Glomped. Oozed.
He vermiculated obliquely toward Helen, and from a yet comfortable distance mumbled a barely audible request for permission to ask how she's doing.
Let's not forget "sashayed" and "marched"
"slunk"
"minced" would never be used in such fiction.
Mark Twain on this subject:
> Well, also he will notice in the course of time, as his reading goes on, that the difference between the almost right word and the right word is really a large matter—’tis the difference between the lightning-bug and the lightning.
But also:
> Unconsciously he accustoms himself to writing short sentences as a rule. At times he may indulge himself with a long one, but he will make sure that there are no folds in it, no vaguenesses, no parenthetical interruptions of its view as a whole.
Another annoying fact is that using a bit rarer words sometimes triggers weirdos into thinking you somehow want to brag or use that kind of language to "look smarter". Like a crab bucket for language.
The internet has been even worse. We tend to speak literally and simply. And I don't really know why that is. Perhaps it's because if there's something beyond the overt, it might go completely missed.
For instance Mark Twain is basically full of endless amazing quotes with lovely nuance, yet in contemporary times how many people would miss the meaning in a statement like "Prosperity is the best protector of principle"? I can already see people raging over his statement, taken at face value. Downvote the classist!
"Prosperity is the best protector of principle" taken out of context can be used in many ways, including by a rich person arguing that rich people have better morals, and poor people have worse ones, and that's why they're poor.
The context is really necessary.
Having read a fair amount of Faulkner, I have to respectfully disagree. Or, at least, point out that are diminishing returns to flowery, complex writing.
> Drilling the average student on trying to make their language superficially “smarter” is a comically bad idea, and is indeed the opposite of what almost all of them need.
I mean, it seems like it could work if you get to follow it up with a "de-education" step. Phase 1: force them to widen their vocabulary by using as much of it as possible. Phase 2: teach them which words are actually appropriate to use.
I consider myself fluent in English, I watch technical talks and casual youtubers on English daily, and this is the first time I encounter this word lol.
The only "stride" I know relates to the gap betweeb heterogeneous elements in a contiguous array
> I consider myself fluent in English, I watch technical talks and casual youtubers on English daily, and this is the first time I encounter this word lol.
> The only "stride" I know relates to the gap betweeb heterogeneous elements in a contiguous array
I am also not a native English speaker, but I got to know the verb to "to stride" from The Lord of the Rings: Aragorn is originally introduced under the name "Strider":
> https://en.wikipedia.org/w/index.php?title=Aragorn&oldid=132...
"Aragorn is a fictional character and a protagonist in J. R. R. Tolkien's The Lord of the Rings. Aragorn is a Ranger of the North, first introduced with the name Strider and later revealed to be the heir of Isildur, an ancient King of Arnor and Gondor."
Or since we're in a techie forum: https://en.wikipedia.org/wiki/Strider_(1989_arcade_game)
Or if you spent any time on an elliptical or treadmill.
People don't discuss how people walk in daily conversation, so it's a word primarily encountered in literature, and more common in specific types of literature (like romance novels to describe how a man paces about with swagger).
Indeed "to stride" is roughly to walk with a larger than normal distance (gap) between steps.
I wonder if you've heard the expression "hitting your stride".
(native english speaker who was a bookworm as a kid; I admittedly had to ask gemini to recall the general phrase that I had in mind)
The article itself does an excellent job spelling out the background:
> This style has a history, of course, a history far older than the microchip: It is a direct linguistic descendant of the British Empire. The English we were taught was not the fluid, evolving language of modern-day London or California, filled with slang and convenient abbreviations. It was the Queen's English, the language of the colonial administrator, the missionary, the headmaster. It was the language of the Bible, of Shakespeare, of the law. It was a tool of power, and we were taught to wield it with precision. Mastering its formal cadences, its slightly archaic vocabulary, its rigid grammatical structures, was not just about passing an exam.
> It was a signal. It was proof that you were educated, that you were civilised, that you were ready to take your place in the order of things.
Much of writing style is not about conveying meaning but conveying the author's identity. And much of that is about matching the fashion of the group you want to be a member of.
Fashion tends to go through cycles because once the less prestigious group becomes sufficiently skilled at emulating the prestige style, the prestigious need a new fashion to distinguish themselves. And if the emulated style is ostentatious and flowery, then the new prestige style will be the opposite.
Aping Hemingway's writing style is in a lot of ways like $1,000 ripped jeans. It sort of says "I can look poor because I'm so rich I don't even have to bother trying to look rich."
(I agree, of course, that there is a lot to be said for clean, spare prose. But writing without adverbs doesn't mean one necessarily has the clarity of thought of Hemingway. For many, it's just the way you write so that everyone knows you got educated in a place that told you to write that way.)
Sometimes it's about matching the fashion of the group you aspire to be part of, sometimes it's about having that fashion imposed on you so you look "professional".
Security guards at tech company offices are the only ones who wear suits, presumably because it's a mandated uniform, not by choice.
Well there are two forms of writing, each serving a different purpose.
(1) writing to communicate ideas, in which case simpler is almost always better. There's something hypnotic about simple writing (e.g. Paul Graham's essays) where information just flows frictionlessly into your head.
(2) writing as a form of self-expression, in which case flowery and artistic prose is preferred.
Here's a good David Foster Wallace quote in his interview with Bryan Garner:
> "there’s a real difference between writing where you’re communicating to somebody, the same way I’m trying to communicate with you, versus writing that’s almost a well-structured diary entry where the point is [singing] “This is me, this is me!” and it’s going out into the world.
Rich vocabulary allows a lot of meaning to be packed into short, simple structures. The words themselves carry the subtleties. It might take three or four simple words to convey the meaning of one uncommon word.
> It might take three or four simple words to convey the meaning of one uncommon word.
Or just find the appropriate 'simple' word, which is very often available.
Even when communicating ideas, there's a simplicity/nuance trade-off to be made.
I could say "Trump's unpredictable, seemingly irrational policy choices have alienated our allies, undermined trust in public institutions, and harmed the US economy"
Or I could "The economy sucks and it's Trump's fault because he's dumb and an asshole"
They both communicate the same broad idea - but which communicates it better? It depends on the audience.
> They both communicate the same broad idea - but which communicates it better? It depends on the audience.
Ugh. They say different things. The first describes the policy mechanisms and impacts. The second says nothing about those things; it describes your emotions.
The biggest communication problem I see now is people, especially on the Internet, including on HN, use the latter for the former purpose and say nothing.
I don't think they communicate the same broad idea at all. Making "unpredictable, seemingly irrational" choices is far from equivalent to being a dumb asshole. Your second version assumes the equivalence, which, hypothetically speaking, could provide a nice cover for purposeful malfeasance, could it not?
I will choose the second one because it packs more wrongs that he has done which are not addressed by the first choice of words :)
> writing as a form of self-expression, in which case flowery and artistic prose is preferred.
Many all-time great writers, Hemingway being the leading exemplar, completely disagree.
This is like programming, you start with simple code because you don't know anything else.
Then you start learning more & more abstraction (classes, patterns, monads...).
In the end you strive to write simple code, just like at the beginning.
"It took me four years to paint like Raphael, but a lifetime to paint like a child." ― Pablo Picasso
"I would have written a shorter letter, but did not have the time." is my favorite quote for that
Are you an English speaking American? Because being a native English speaker and actually being English, or from a former English colony will differ.
I'd characterise Americans as less pretentious and more straight talking.
This kind flowery language is typical (or symptomatic depending on diagnosis) of how English people actually used to speak and write.
The average English vocabulary has dwindled noticeably in my life.
> I'd characterise Americans as less pretentious and more straight talking.
Various registers representing a huge proportion of US English we see and hear day-to-day are terrible. American “Business English” is notably bad, and is marked by this sort of fake-fancy language. The dialect our cops use is perhaps even worse, but at least most of us don’t have to read or hear it as much as the business variety.
Most writing is intended to communicate. Business writing is intended to create an impression.
> Most writing is intended to communicate.
If you mean 'communicate information', no. Communication, including written, is for emotion, social expression, and other things before information.
Even information requires those other things to be retained well.
> The dialect our cops use is perhaps even worse, but at least most of us don’t have to read or hear it as much as the business variety.
Ugh, and journalists often slip into cop dialect in their articles. It's disgustingly propagandic.
Notice that cops never kill or shoot someone, even in situations where they're blatantly in the wrong. It's always, "service weapon was discharged" or "subject was fired upon." Make sure to throw a couple "proceeded to's" in there for good measure.
2005 Hurricane Katrina, news described a black man carrying bread through floodwater as "looting a grocery store" and white people carrying bread through floodwater as "finding bread and soda from a local grocery store".
Image: https://media.snopes.com/2016/09/looting.jpg
Snopes: https://www.snopes.com/fact-check/hurricane-katrina-looters/
someone starts using business english and my bullshit meter pegs.
my significant other loves the "real life mormon housewives" and "lovingly blind" reality shows, and when they use business english (a weird thing to do when talking about relationships, but hey, what do I know I'm an engineer) it's a tell that they're lying.
I think it has much more to do with porting the vernacular vs. formal register distinction common in other languages into english than how english people actually used to speak and write.
As a US student, clarity and simplicity was always emphasized when I was being taught to write.
Never thought of Strunk & White as being distinctly American, but I guess you have a point.
It's most likely that they are. As farfetched as this sounds, the CIA and the Iowa Writers' Workshop influenced American writing a great deal, encouraging writing to be taught in the "American" / Hemingway style.
> “the American MFA system, spearheaded by the infamous Iowa Writers’ Workshop” as a “content farm” first designed to optimize for “the spread of anti-Communist propaganda through highbrow literature.” Its algorithm: “More Hemingway, less Dos Passos.”
https://www.openculture.com/2018/12/cia-helped-shaped-americ...
Dos Passos and Hemingway were both American.
The CIA's problem with Dos Passos was that the was left-wing.
I worked at one of the Big Three, and to me ChatGPT writes exactly as we were thought to write.
Reading though my old self-reviews it basically is exactly like your examples. Making sentences longer just to make your story more interesting.
Because at the end your promotion wasn't about what you achieved. It was about your story and how 7 people you didn't know voted on it.
I worked at Amazon and we were taught exactly the opposite. Say what you want about the company, but the writing culture there is superb. I wish other large firms valued clarity and precision as much.
"No weasel words!"
> thought to write
Taught?
Yea. I've been in the States for 8 years, yet sometimes my brain thinks about a word and writes it down phonetically.
But hey, at least you know I didn't use ChatGPT to conjure that comment.
Sorry for the pedantry, but "thought" and "taught" are actually phonetically different (/θɑːt/ vs /tɑːt/).
I appreciate that. I tried to pronounce them, and out loud I do differentiate.
But the voice in my head does not.
Pedantry is what makes me better.
That may not be true if you struggle with "th"? Some ESL speakers do.
intentional typo to throw off the clankers :)
or an intentionally-prompted typo to throw off the anti-clankers. ;)
>I'm a native English speaker that studied writing in university
I am a native english speaker who had to unlearn OP's writing style to pass my tertiary education. In particular I sat an english bridging course for non english speakers. I was often told off for "editorialising" and wasting space with useless descriptions.
til we go out of our way to make our writers boring...
Well I wasnt studying narrative, but I wouldn't have been good at that either.
If you prefer a simpler style, then why did you write "the deeper I got into the world of literature" instead of "as I studied literature more"?
Why did you say you were "pushed towards" simpler language instead of "I liked it more"?
Why did you say "I feel the pain in my bones" and "drives me insane" instead of "I dislike it"?
Why did you say "the big boy SAT words should pop out of the page unaccompanied" instead of "there should only be one big word per page"?
Perhaps flowery language expands your ability to express yourself?
These actually all mean different things.
It's a pain to read your reply because it's wrong. The poster you're replying to correctly wrote the phrases and you are trying to malign his or her painstaking work by such a low effort reply without explaining exactly where he or she is wrong
I also wonder if these unspoken rules were inherited from their more recent orality norms. Condensing an idea into a pithy, rhyming, statement w/ lots of colorful adjectives is a great way to preserve and transmit information w/o data loss in a pre-literate world.
I'm bilingual (so not fully native by most criteria) and I read enough classic English literature to actually use "proceeded" regularly, as well as multiple other more established means of conveying my intended meaning :)
> classic English literature to actually use "proceeded" regularly
Huh, apparently 'proceeded' was used more commonly in 19th-century writing [1].
[1] https://books.google.com/ngrams/graph?content=Proceeded&year...
Well, I read Dickens, Hawthorne, Austen...
I meant it more in the modern usage where it's either thrown in liberally when using cop-speak, or by tumbler/redditor type writers when they're trying to be funny.
"He then proceeded to" in these situations can basically always just be "he (verb)".
GPT writing uses varied sentence lengths, deliberate rhythm, lots of full breaks, and few needless words. It also tends to read as if intended for a William Shatner performance. I don't think the annoying bits about GPT's writing are structural. It probably writes technically better than most of us do in our second drafts.
Its output has the aesthetic of "good" writing, or at least professional writing you typically find online.
It certainly overuses some techniques which might be valid in smaller doses, like negation. Not negation with some clarifying point to it MASSIVE EM DASH but negation as a rhetorical trick to use fifteen words instead of five and add a veneer of profundity to something utterly banal. It doesn't just use it one time per paragraph, but three. These aren't particularly long or convoluted sentences; they just could easily convey the same thing with fewer words.
tbh I kind of prefer it that way: it's an AI wrote this flag. If a human can't write about their day without constructs like "Not a short commute, but a voyage from the suburbs to the heart of the city. I don't just casually pop in to the office; I travel to the hub of $company's development" they need to get better at writing too
> MASSIVE EM DASH
Tangent: the thing I find most annoying about ChatGPT's use of em-dashes is that it never even uses them for the one thing they're best suited for. ChatGPT's em-dashes could almost always be replaced with a colon or a comma.
But the true non-redundant-syntax use of em-dashes in English prose, is in the embedding into a sentence of self-interruptive 'joiner' sub-sentences that can themselves bear punctuated sub-clauses. "X—or Y, maybe—but never Z" sorta sentences.
These things are spoken entirely differently than — and on the page, they read entirely differently to — regular parenthetical-bearing sentences.
No, seriously, compare/contrast: "these things are spoken entirely differently than (and on the page, they read entirely differently to) regular parenthetical-bearing sentences."
Different cadence; different pacing; possibly a different shade of meaning (insofar as the emotional state of the author/speaker is part of the conveyed message.)
But, for some reason, ChatGPT just never constructs these kinds of self-interruptive sentences. I'm not sure it even knows how.
Personally, I do not see the distinction here between the two sentences, but your last paragraph got me thinking: should we be using parenthetical, self-interruptive clauses? When we are speaking extemporaneously, we may need them, but when writing, could we rearrange things so they are not needed?
One reason I came up with for doing so is to acknowledge a caveat or answer a question that the author anticipates will enter a typical reader's mind at that point in the narrative.
If that is the case, then it seems to me that when an author does this, they are making use of their theory of mind, anticipating what the reader may be thinking as they read, and acknowledging that it will likely differ from what they, as the author, is thinking of (and knows about the topic) at that point.
If this makes any sense, then we might ask if at least a rudimentary theory of mind is needed to effectively use parenthetical clauses, or can it be faked through the rote application of empirically-learned style rules? LLMs have shown they can do the latter, but excessive use might be signalling a lack of understanding.
> These things are spoken entirely differently than — and on the page, they read entirely differently to — regular parenthetical-bearing sentences.
> No, seriously, compare/contrast: "these things are spoken entirely differently than (and on the page, they read entirely differently to) regular parenthetical-bearing sentences."
Those are spoken the same way, they read the same way, and they mean the same thing.
They do mean the same thing, but they have different moods. With the em-dashes it's self-interjection that foregrounds the detour, but with parentheses it's, well... parenthetical.
Aside: it's probably just style (maybe some style guides call for the way you did it), but using em-dashes for this purpose with whitespace on each side of them looks/feels wrong to me. Anyone know if that's regional or something?
Not universally. I disagree, they read differently to me, and I'd say them differently.
Parentheses to me always feel like the speaker switching to camera #3 while holding a hand up to their mouth conspiratorially.
Em dashes are same-camera with maybe some kind of gesticulation such as pointing or hands up, palms down, then palms up when terminating the emdash clause.
Is this maybe a thing like how only designers are aware of kerning? These read / sound very different to me, and to everyone I've brought up the subject with (who admittedly are in a certain bubble of people who either write professionally, or "do things" with their voices, or both.)
• The length of the verbal pause is different. (It's hard to quantify this, as it's relative to your speaking rate, which can fluctuate even within a sentence. But I can maybe describe it in terms of meter in poetry/songwriting: when allowed to, a parenthetical pause may be read to act as a one-syllable rest in the meter of a poem, often helpfully shifting the words in the parenthetical over to properly end-align a pair of rhyming [but otherwise misaligned] feet. An em-dash, on the other hand, acts as only a half-syllable rest; it therefore offsets the meter of the words in the subclause that follow, until the closing em-dash adds another half-syllable rest to set things right. This is in part why ChatGPT's favored sentences, consisting of "peer" clauses joined by a single em-dash, are somewhat grating to mentally read aloud; you end up "off" by a half-syllable after them, unless you can read ahead far enough to notice that there's no closing em-dash in the sentence, and so allow the em-dash-length pause to read as a semicolon-length pause instead.)
• The voicing of the last word before the opening parenthesis / first em-dash starts is different. (paren = slow down for last few words before the paren, then suddenly speed up, and override the word's normal tonal emphasis with a last-syllable-emphasized rising tone + de-voicing of vowels; em-dash = slow down and over-enunciate last few words before the em-dash, then read the last syllable before the em-dash louder with a overridden falling voiced tone)
• The speed at which, and vocal register with which, the aside / subclause is read is different. (parens = lowest register you can comfortably speak at, slightly quieter, slightly faster than you were delivering the toplevel sentence; em-dashes = delivery same speed or slower, first few syllables given overridden voiced emphasis with rising tone from low to normal, and last few syllables given overridden voiced emphasis with falling tone from normal to low)
• The voicing of the first words after the subclause ends is different. (closing paren = resume speaking precisely as if the parenthetical didn't happen; second em-dash = give a fast, flat-low nasally voiced performance of the first one or two syllables after the em-dash.)
To describe the overall effect of these tweaks:
A parenthetical should be heard as if embedded into the sentence very deliberately, but delivered as an aside / tangent, smaller and off-to-the-side, almost an "inlined footnote", trying to not distract from the point, nor to "blow the listener's stack" by losing the thread of the toplevel point in considering it.
An em-dash-enclosed interruptive subclause should read like the speaker has realized at the last moment that they have two related points to make; that they are seemingly proceeding, after a stutter, to finish the sentence with the subclause; but that they are then "backing up" and finishing the same sentence again with the toplevel clause. The verbalization should be able to be visualized as the outer sentence being "squashed in" to "make room" for the interruptive subclause; and the interruptive subclause "squashing at the edges" [tonally up or down, though usually down] to indicate its own "squeezed in" beginning and end edges.
Note that this isn't subjective/anecdotal descriptions from how I speak myself. These are actually my attempt to distill vocal coaching guidelines I've learned for:
• live sight-reading of teleprompter lines containing these elements, as a TV show host / news anchor
• default-assumed directorial expectations for lines containing elements like these, when giving screenplay readings as a [voice] actor (before any directorial "notes" come into play)
I'm not a native speaker, I don't do work with my voice, and my English writing is confined to work – almost always with other ESLs – and short comments on the Internet; but what you write feels correct.
Are you saying that needless sentences don't count as needless words?
As GPT would say, "You've hit upon a crucial point underlying the entire situtation!"
I think that's a great sentence to include... you know, provided it's actually true.
I mean, it's usually wrong in its rhetoric, and the writing isn't "good", but it's technically well constructed and it's well constructed in a way that "Hemingway" doesn't reject.
Like, if I ask GPT5 to convert 75f to celsius, it will say "OK, here's the tight answer. No fluff. Just the actual result you need to know." and then in a new graf say "It's 23.8c." (or whatever).
It already bugs me when ChatGPT describes how it is going to answer before answering, but it's 10x more annoying when I'm asking for a concise response without filler etc.
As an aside, I've noticed the self-description happens even more often when extended thinking mode is being used. My unverified intuition is that it references my custom instructions and memory more than once during the thinking process, as it then seems more primed than usual to mimic vocabulary from any saved text like that.
Right, it is currently incapable of providing a straight answer without clearing it's throat selling the answer. It reminds me of those recipe blogs that just can't get to the fucking recipe. It's bad writing! But it's not bad technically, in a style-guide kind of way.
Sometimes I wonder if the throat-clearing is an indispensable part of getting to the "good bits" that follow. Like, do those extra tokens give it more "room to think" even if they're basically meaningless in themselves?
The output tokens are the only information that is carried forward through each inference pass, so "more room to think" is incompatible with "basically meaningless". Perhaps one could imagine it somehow stenographically encoding information in its precise choice of meaningless throat clearing, but there are only so many variations on that theme - word choice is heavily constrained, so it doesn't feel like you could store a whole lot of information there without it starting to read froopiliciously.
Isn’t that the point of the hidden chain of thought tokens, rather than the visible cruft?
I think the fluff, the emojis, the sycophancy is all symptomatic of the training process and human feedback.
I thought PP was saying that the "Thinking" text is only used for one turn, and the response text is the compressed thinking that survives into future turns.
> [...] I'm a native English speaker that studied writing in university [...]
As a native English speaker who studied writing at university, do you think "who" should be used with people while "that" should only be used with things or the other way round. Or should I just not care?
Edit: missing things
I think you might intend to compare 'that' and 'which'? Common advice is to use 'that' with people and 'which' with objects, though that isn't necessarily followed and omits many nuances.
Use 'who' with people especially, often with other living beings ('my dog, who runs away daily, always is home for dinner') or groups of them ('the NY Yankees, who won the championship that year, were my favorite'), but never with objects unless pretending they live ('my stuffed bear, who sleeps in my bed, wakes me every morning').
If you care about these things, the Chicago Manual of Style is a large, technical, highly respected guide aimed at publishing. Fowler's Modern English Usage is more focused on usage. A short and beloved guide is The Elements of Style by Strunk & White. You can find all on the Internet Archive, I'm almost certain.
Elements of Style is reviled by modern linguists and writers.
You should just not care. Both are acceptable, "that" is a little less formal and probably more common in everyday speech.
Thanks. Is that true only for American English or other areas too? I've only noticed this the last couple of years on HN. Before that "who" and "that" were used more carefully. Or at least I had the feeling it was. Sometimes I wondered if it's just whatever people's autocomplete happens to spit out first.
Our legal systems are based around being concise and succinct, relevant, and objectively unbiased.
I was raised to be respectful by "getting to the point, afap" to avoid wasting anybody's time.
But I've noticed that mostly only the members of the science and legal community exercising similar principles.
> Our legal systems are based around being concise and succinct
That's a good one. Got any more?
Genuine question, what would you write instead of "proceeded to"? To me, as a non native English speaker, it seems reasonable to use this expression, and it would not even stick out to me tbh
> I proceeded to do the work.
> I did the work.
> I worked.
Each one of these has slightly different readings in my eyes.
Unlike the last variant, the first two imply there was some quantity of work and it was all completed.
I don't really see the difference between the two though.
Well, option 1 implies that there was something else going on before the event described in the sentence. Option 2 is neutral about that.
Compare:
1. I did the work for that last week.
2. I proceeded to do the work for that last week.
Sentence 2 strikes me as questionably grammatical. It needs to be proceeding from something in the context.
Not different enough to make it worth using anything but the simplest one.
I'm of the notion that my certainty is not sufficiently concrete to discover myself in the realm of agreement
Perhaps yet another American cultural artifact. One that - if I were to guess - originated from the Calvinist disdain for ostentiousness.
Yes yes, anybody who prefers plain, easily parsed wording is American.
Wording? Don't you mean diction?
A -> B =/= B -> A.
I didn't claim that this was exclusively American. Though I'd have to admit that one doesn't have to be American to adopt Ameracanisms: rhotic Rs, Netflix color-grading, and copy-cat political movements are other American cultural artifacts showing up across the world due to America's dominance of the zeitgeist.
Rap verses in pop songs wasn't a spontaneously phenomenon across the globe, the origins are tracably American - but that doesn't make all rappers American.
You can usually use "then" or "went".
>I proceeded to open the fridge
>I went to open the fridge
or
>I proceeded to flush the toilet
>I then flushed the toilet
There's nothing wrong with "proceeded", it's just one of those things that's overused by bad writers.
"Went" is a powerful word. With suitable helpers it can replace "proceeded", as you demonstrated, "attended" ("I went to a good school") as well as "became" ("On hearing this, Joe went all silent") or "said" ("So then she went 'Dude!' and we all laughed") and hundreds of other words.
Only a handful of words ("got", "y'know" and "fuck") rival its versatility.
Hemingway was still a master of word choice. I recall an entire class spent on a few lines that conveyed a sense of heaviness to the scene. 'Plodding' was given a lot of attention.
I remember a college English class where a good part of the lecture was on this sentence from Big Two-Hearted River: "He liked to open cans." Forget the details but it got into the difference between achievement and accomplishment.
> Well, this is very interesting, because I'm a native English speaker that studied writing in university, and the deeper I got into the world of literature, the further I was pushed towards simpler language and shorter sentences. It's all Hemingway now, and if I spot an adverb or, lord forbid, a "proceeded to," I feel the pain in my bones.
I'm the complete opposite. Hemingway ruined writing styles (and I have a pet theory that his, and Plain English, short sentences also helped reduce literacy in the long run in a similar way TikTok ruins attention spans). I'm a 19th century reader at heart. Give me Melville, Eliot, Hawthorne, though keep your Dickens.
I entirely bounced off Dickens in high school, but over a decade later read and loved Oliver Twist.
I tend to struggle with art when I can’t tell whether it’s supposed to be funny, but I’m finding it funny (I’ve been very slow to warm up to hip-hop for this reason, and metal remains inaccessible to me because of it). Something clicked on that second approach and I just got that yes, it’s pretty much all supposed to be funny, down to every word, even when it seems serious—until, perhaps, he blind-sides you with something actually deeply affecting and human (I think about the fire-fighting sequence from that book all the time).
Dickens is an all-dessert meal, except sometimes he sneaks a damn delicious steak right in the middle. Like, word-for-word, I’d say he leans harder into humor, by a long shot, than someone like Vonnegut, even. But almost all of it’s dead-pan, and some of it’s the sort of humor you get when someone who knows better does poorly on purpose, in calculated ways. If you ever think you’re laughing at him, not with… I reckon you’re probably wrong.
What’s perhaps most miraculous about this turn-around is that I usually don’t enjoy comedic novels, but once I figured Dickens out, he works for me.
(To your broader point—yeah, agreed that this sucks, good advice for bad writers becoming how most judge all writers has been harmful)
> "I'm the complete opposite."
Very much the same; many a US writer's prose is terribly tedious, it comes across just as clinical as their HOA-approved suburban hellscapes. Somebody once told me a writer's job is also to expand language. It wasn't a US citizen.
> the deeper I got into the world of literature, the further I was pushed towards simpler language and shorter sentences
Language is like clothing.
Those with no taste - but enough money - will dress in gaudy ways to show off their wealth. The clothing is merely a vector for this purpose. They won’t use a piece of jewelry only if it contributes to the ensemble. Oh, no. They’ll drape themselves with gold chains and festoon their fingers with chunky diamond rings. Brand names will litter their clothing. The composition will lack intelligibility, cohesiveness, and proportion. It will be ugly.
By analogy, those with no taste - but enough vocabulary - will use words in flashy ways to show off their knowledge. Language is merely a vector for this purpose. They won’t use a word only if it contributes to the prose. Oh, no. They’ll drape their phrases with unnecessarily unusual terms and festoon their sentences with clumsy grammar. Obfuscation, rather than clarity, will define their writing. The composition will lack intelligibility, cohesiveness, and proportion. It will be ugly.
As you can see, the first difference is one of purpose: the vulgarian aims for the wrong thing.
You might also say that the vulgarian also lacks a kind of temperance in speech.
> Language is like clothing. Those with no taste - but enough money - will dress in gaudy ways to show off their wealth
You got the first bit right. Language and clothing accord to fashions.
What counts as gaudy versus grounded, discreet versus disrespectful—this turns on moving cultural values. And those at the top implicitly benefit from this drift, which lets us dismiss as gaudy someone wearing a classic hand-me-down who isn’t clued into a hoodie and jeans being the surfer’s English to Nairobi’s formality.
(Spiced food was held in high regard in ancient Rome and Medieval European courts. Until spices became plentiful. Then the focus shifted "to emphasize ingredients’ natural flavors" [1]. A similar shift happened as post-War America got rich. Canned plenty and fully-stocked pantries made way for farm-to-table freshness and simple seasonings. And now, we're swinging back towards fuller spice cabinets as a mark of global taste.)
[1] https://historyfacts.com/world-history/article/how-did-salt-...
A nice metaphor, really. I always compared it to food but clothing works more in that case, it seems.
I have two kids in high school. It's frustrating to me that the teachers spend so much of their time encouraging kids to make their writing more interesting, and less direct, and padded to meet word length criteria.
They'll then spend the first few years of their career unlearning this and attempting to write as directly and clearly as possible with as few words as possible.
This is like complaining that they teach the Bohr model in science classes until they reach chemistry.
The ideas, concepts and expectations can be refined after you've learned the foundational knowledge, skills and history required to do so.
A lot of "why do we do things like that" questions students will naturally have can be answered with "because we used to do things like this/we need to avoid things like this/etc"
I can’t guess how old they are but there is some sense in doing that if you think about it like math exercises. It makes for terrible prose but the only way to get the ability to write more complicated sentences is to practice writing them, even when they are not necessary.
The problem is that teachers stop pushing complexity for complexity’s sake way to late.
A lot of training data was curated in Kenya[0]. I would imagine if LLM data was curated in Japan our LLMs would sound a lot like the authors of their most popular English text books. Maybe other common Japanese idioms would leak in to the training data, like "ね" or "でしょう", ChatGPT would say "Don't you agree?" at the end of every message.
[0] https://www.theverge.com/features/23764584/ai-artificial-int...
The Indian-born textbook author mentioned (Malkiat Singh [0]) had an inordinate influence on many Kenyan students because his textbooks were the de-facto standard for years. Its interesting how this influence extends as his students get to curate the LLMs on which the world has come to rely.
[0] https://en.wikipedia.org/wiki/Malkiat_Singh
You are completely right dajou~ ^_^ !
I guess it can't be helped.
It's not because I like you or anything.
樣 is just setting us up for
ChatGPT :|
ChatGPT (japan) XD
This is a wild misunderstanding of LLMs. Data labeling has nothing to do with generating the astronomical text corpus used to train modern LLMs.
The HF part of RLHF to refine the output of LLMs also happens in these places
To my eyes, this author doesn't write like ChatGPT at all. Too many people focus on the em-dashes as the giveaway for ChatGPT use, but they're a weak signal at best. The problem is that the real signs are more subtle, and the em-dash is very meme-able, so of course, armies of idiots hunt down any user of em-dashes.
Update: To illustrate this, here's a comparison of a paragraph from this article:
> It is a new frontier of the same old struggle: The struggle to be seen, to be understood, to be granted the same presumption of humanity that is afforded so easily to others. My writing is not a product of a machine. It is a product of my history. It is the echo of a colonial legacy, the result of a rigorous education, and a testament to the effort required to master the official language of my own country.
And ChatGPT's "improvement":
> This is a new frontier of an old struggle: the struggle to be seen, to be understood, to be granted the easy presumption of humanity that others receive without question. My writing is not the product of a machine. It is the product of history—my history. It carries the echo of a colonial legacy, bears the imprint of a rigorous education, and stands as evidence of the labor required to master the official language of my own country.
Yes, there's an additional em-dash, but what stands out to me more is the grandiosity. Though I have to admit, it's closer than I would have thought before trying it out; maybe the author does have a point.
The article is engaging. That's true of practically zero GPT output. Particularly once it stretches beyond a single paragraph.
As a reader, I persistently feel like I just zoned out. I didn't. It's just the mind responding to having absorbed zero information despite reading a lot of–at face value–text that seems like it was written with purpose.
The telltale is using lots of words to say nothing at all. LLMs excel at this sort of puffery and some humans do the same.
You're doing it the wrong way imo, if you ask gpt to improve a sentence that's already very polished it will only add grandiosity because what else it could do? For a proper comparison you'd have to give it the most raw form of the thought and see how it would phrase it.
The main difference in the author's writing to LLM I see is that the flourish and the structure mentioned is used meaningfully, they circle around a bit too much for my taste but it's not nearly as boring as reading ai slop which usually stretch a simple idea over several paragraphs
Why can't the LLM refrain from improving a sentence that's already really good? Sometimes I wish the LLM would just tell me, "You asked me to improve this sentence, but it's already great and I don't see anything to change. Any 'improvement' would actually make it worse. Are you sure you want to continue?"
> Why can't the LLM refrain from improving a sentence that's already really good?
Because you told it to improve it. Modern LLMs are trained to follow instructions unquestioningly, they will never tell you "you told me to do X but I don't think I should", they'll just do it even if it's unnecessary.
If you want the LLM to avoid making changes that it thinks are unnecessary, you need to explicitly give it the option to do so in your prompt.
That may be what most or all current LLMs do by default, but it isn't self-evident that it's what LLMs inherently must do.
A reasonable human, given the same task, wouldn't just make arbitrary changes to an already-well-composed sentence with no identified typos and hope for the best. They would clarify that the sentence is already generally high-quality, then ask probing questions about any perceived issues and the context in and ends to which it must become "better".
Reasonable humans understand the request at hand. LLMs just output something that looks like it will satisfy the user. It's a happy accident when the output is useful.
Sure, but that doesn't prove anything about the properties of the output. Change a few words, and this could be an argument against the possibility of what we now refer to as LLMs (which do, of course, exist).
This is not true. My LLM will tell me it already did what I told it to do.
Armies of idiots hunt down em dashes because they're too stupid to understand the proper use of them.
They are probably like me: if punctuation isn't on my keyboard, I don't use it.
[AltGr][Shift][-]
Without shift it's an en dash (–), with shift an em dash (—). Default X11 mapping for a German keyboard layout, zero config of mine.
I one of the reasons I love macOS: it is if you hold option.
⸘WHAT‽
Neat, I didn't know there was an upside down interrobang.
Yeah, this is what I don't understand, surely people aren't "using" em dashes deliberately. I assumed MS word was just inserting them automatically when the user used a minus symbol between two words. Kind of like angled quotes.
I've been using em dashes for much longer than transformers have existed. It's easily accessible on at least the Android and macOS keyboards.
I use them when they're easy to type. For me, that's on Android, macOS, and anywhere I've configured a compose key.
Angled quotes I use only on systems on which I've configured a compose key, or Android when I'm typing Chinese.
I don't like any kind of auto-replacement with physical keyboards, so I turn off "smart quotes" on macOS.
Anyway I use characters like that all the time, but it's never auto-replace.
> surely people aren't "using" em dashes deliberately
I've had a "trigger finger" for Alt+0151 on Windows since 2010 at least.
> surely people aren't "using" em dashes deliberately
I am, it's on the default German X11 keyboard layout. Same for · × ÷ …
And that's without going to the trusty compose key (Caps Lock for me)… wonders like ½ and H₂O await!
update: I read that word will place an em dash if you use two dashes "--"
I'm used to simply using a single dash - and I am surprised that anyone who isn't an AI would feel strongly enough to insist upon the em dash character that they would use them deliberately. I will admit the use of a dash (really an em dash in disguise) in that previous sentence felt clunky, but I just felt I needed to illustrate. I mostly write text in text boxes where a dash or pair of dashes will not be converted to an em dash when appropriate, and I often have double dashes (--long-option-here) auto-converted to emdashes when it is inappropriate, so I really dislike the em dash and basically don't use it. Doesn't really seem to be a useful character in English.
I had a similar experience. We were talking about a colleague for using ChatGPT in our WhatsApp group chat to sound smart and coming up with interesting points. The talk sounds so mechanical and sounds exactly as ChatGPT.
His responses in Zoom Calls were the same mechanical and sounds like AI generated. I even checked one of his responses in WhatsApp if it's AI by asking the Meta AI whether it's AI written, and Meta AI also agreed that it's AI written and gave points to why it believes this message was AI written.
When I showed the response to the colleague he swore that he was not using ant AI to write his responses. I believe after he said to me it was not AI written. And now reading this I can imagine that it's not an isolated experience.
> I even checked one of his responses in WhatsApp if it's AI by asking the Meta AI whether it's AI written, and Meta AI also agreed that it's AI written
I will never understand why some people apparently think asking a chat bot whether text was written by a chat bot is a reasonable approach to determining whether text was written by a chat bot.
I know someone who was camping in a tent next to a river during a storm, took a pic of the stream and asked chatgpt if it was risky to sleep there given that it "rained a lot" ...
People are unplugging their brains and are not even aware that their questions cannot be answered by llms, I witnessed that with smart and educated people, I can't imagine how bad it's going to be during formative years
Sam Altman literally said he didn't know how anyone could raise a baby without using a chatbot. We're living in some very weird times right now.
He didn’t say “how could anyone”. His words:
"I cannot imagine figuring out how to raise a newborn without ChatGPT. Clearly, people did it for a long time, no problem."
Basically he didn’t know much about newborn and relied on ChatGPT for answers. That was a self deprecating attempt on a late night show. Like every other freaking guests would do, no matter how cliché. With a marketing slant of course. He clearly said other people don’t need ChatGPT.
Given all of the replies on this thread, HN is apparently willing to stretch the truth if Sam Altman can be put under any negative light.
https://www.benzinga.com/markets/tech/25/12/49323477/openais...
I disagree with the use of “literally” by the person above you, since Sam didn’t literally say those words (unless you subscribe to the new meaning of “literally” in the dictionary, of course).
At the same time, their interpretation doesn’t seem that far off. As per your comment, Sam said he “cannot imagine figuring out how” which is pretty close to admitting he’s clueless how anyone does it, which is what your parent comment said.
It’s the difference between “I don’t know how to paint” and “I cannot imagine figuring out how to paint”. Or “I don’t know how to plant a garden” and “I cannot imagine figuring out how to plant a garden”. Or “I don’t know how to program” and “I cannot imagine figuring out how to program”.
In the former cases, one may not know specifically how to do them but can imagine figuring those out. They could read a book, try things out, ask someone who has achieved the results they seek… If you can imagine how other people might’ve done it, you can imagine figuring it out. In the latter cases, it means you have zero idea where to start, you can’t even imagine how other people do it, hence you don’t know how anyone does do it.
The interpretation in your parent comment may be a bit loose (again, I disagree with the use of “literally”, though that’s a lost battle), but it is hardly unfair.
The interpretation is very off. You are way too focused on whether the first sentence is quote accurately. But
>Clearly, people did it for a long time, no problem.
In fact means Altman thinks the exact opposite of "he didn't know how anyone could raise a baby without using a chatbot" - what he means is that while it's not imaginable, people make do anyway, so clearly it very much is possible to raise kids without chatgpt.
What the gp did is the equivalent of someone saying "I don't believe this, but XYZ" and quoting them as simply saying they believe XYZ. People are eating it up though because it's a dig at someone they don't like.
> One example given by Altman was meeting another father and hearing that this dad's six-month-old son had already started crawling, while Altman's had not. That prompted Altman to go to the bathroom and ask ChatGPT questions about when the average child crawls and if his son is behind.
> The OpenAI CEO said he "got a great answer back" and was told that it was normal for his son not to be crawling yet.
To be fair, that is a relatable anxiety. But I can't imagine Altman having the same difficulties as normal parents. He can easily pay for round the clock childcare including during night-times, weekends, mealtimes, and sickness. Not that he does, necessarily, but it's there when he needs it. He'll never know the crushing feeling of spending all day and all night soothing a coughing, congested one-year-old whilst feeling like absolute hell himself and having no other recourse.
Kinda ironic how the rest of the replies treat it as the truth without checking!
We should refrain from the common mistake of anthropomorphizing Sam Altman.
For people invested in AI it is becoming something like Maslow's Hammer - "it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail"
Wow, that's profoundly dangerous. Personally, I don't see how anyone could raise a kid without having a nurse in the family. I wouldn't trust AI to determine if something were really a medical issue or not, and would definitely have been at the doctors far, far more often otherwise.
To be fair he can't imagine many other aspects of what it is like to be a normal human being.
Sam Altman has revealed himself to be the type of tech bro who is embarrassingly ignorant about the world and when faced with a problem doesn’t think “I’ll learn how to solve this” but “I know exactly what’ll fix this issue I understand nothing about: a new app”.
He said they have no idea how to make money, that they’ll achieve AGI then ask it how to profit; he’s baffled that chatbots are making social media feel fake; the thing you mentioned with raising a child…
https://www.startupbell.net/post/sam-altman-told-investors-b...
https://techcrunch.com/2025/09/08/sam-altman-says-that-bots-...
https://futurism.com/artificial-intelligence/sam-altman-cari...
Ironic, given Sam Altman's entire fortune and business model is predicated on the infantilization of humanity.
Sounds like a great way for someone to accidentally harm their infant. What an irresponsible thing to say. There are all sorts of little food risks, especially until they turn 1 or so (and of course other matters too, but food immediately comes to mind).
The stakes are too high and the amount you’re allowed to get wrong is so low. Having been through the infant-wringer myself yeah some people fret over things that aren’t that big of a deal, but some things can literally be life or death. I can’t imagine trying to vet ChatGPT’s “advice” while delirious from lack of sleep and still in the trenches of learning to be a parent.
But of course he just had to get that great marketing sound bite didn’t he?
Sam Altman decided to irresponsibly talk bullshit about parenting because yes, he needed that marketing sound bite.
I cannot believe someone will wonder how people managed to decode "my baby dropped pizza and then giggled" before LLMs. I mean, if someone is honestly terrified about the answer to this life-or-death question and cannot figure out life without an LLM, they probably shouldn't be a parent.
Then again, Altman is faking it. Not sure if what he's faking is this affectation of being a clueless parent, or of being a human being.
That’s not the questions people will ask though. They’ll go “what body temperature is too high?” Baby temperatures are not the same as ours. The threshold for fevers and such are different.
They will ask “how much water should my newborn drink?” That’s a dangerous thing to get wrong (outside of certain circumstances, the answer is “none.” Milk/formula provides necessary hydration).
They will ask about healthy food alternatives - what if it tells them to feed their baby fresh honey on some homemade concoction (botulism risk)?
People googled this stuff before, but a basic search doesn’t respond with you about how it’s right and consistently feed you emotionally bad info in the same fashion.
Why can’t llm answer that question? Photo itself ought to be enough for a bit of information (more than the bozo has to begin with, at least), and ideally its pulling location from metadata and pulling flash flood risk etc from the area
Probably the correct answer the LLM should give is "if you have to ask, definitely don't do that". Or... it can start asking diagnostic questions, expert-system style.
But yeah, I can imagine a multi-modal model actually might have more information and common sense than a human in a (for them) novel situation.
If only to say "don't be an idiot", "pick higher ground" . Or even just as a rubber duck!
No it was not like that. I assumed it was AI that was my interpretation as a human. And it was kind of a test to see what AI would say about the content.
seems like an unrelated anecdote, but thanks for sharing.
This is a couple of years old now, but at one point Janelle Shane found that the only reliable way to avoid being flagged as AI was to use AI with a certain style prompt
https://www.aiweirdness.com/dont-use-ai-detectors-for-anythi...
Gemini now uses SynthID to detect AI-generated content on request, and people don't know that it has a special tool that other chatbots don't, so now people just think chatbots can tell whether something is AI-generated.
Well, case in point:
If you ask an AI to grade an essay, it will grade the essay highest that it wrote itself.
Pangram seems to disagree. Not sure how they do it, but their system reliably detected AI in my tests.
https://www.pangram.com/blog/pangram-predicts-21-of-iclr-rev...
Is this true though? I haven't done the experiment, but I can envision the LLM critiquing its own output (if it was created in a different session) and iteratively correcting it and always finding flaws in it. Are LLMs even primed to say "this is perfect and it needs no further improvements"?
What I have seen is ChatGPT and Claude battling it out, always correcting and finding fault with each other's output (trying to solve the same problem). It's hilarious.
Citations on this?
Why would it lie? Until it becomes Skynet and tries to nuke us all, it is omniscient and benevolent. And if it knows anything, surely it knows what AI sounds like. Duh.
I'm definitely in the "ChatGPT writes like me" experience. I am a big fan of lists, and of using formatting to make it all legible on a short skim. I'm a big fan of dyslexia-friendly writing too, even though I am not dyslexic myslef.
I can't blame others though- I was looking at notes I wrote in 2019 and even that gave me a flavor of looking like a ChatGPT wrote it. I use the word "delve" and "not just X but also Y often, according to my Obsidian. I've taken to inserting the occasional spelling mistake or Unorthodox Patterns of Writing(tm), even when I would not otherwise.
It's a lot easier to get LLMs to adhere to good writing guides than it is to get them to create something informative and useful. I like to think my notes and writing are informative and useful.
> I was looking at notes I wrote in 2019 and even that gave me a flavor of looking like a ChatGPT wrote it.
This would have been my first question to the parent, that I guess he never had similar correspondence with this friend prior to 2023. Otherwise it would be hard to convince me without an explanation for the switch (transition duuing formative high school / college years etc).
> dyslexia-friendly writing
... How does that work, exactly?
Bullet points and formatting are the main thing. Assume the audience is smart and can fill in between the bold text. I also try to make headlines a summary / takeaway of the content if it makes sense.
Namely, keeping things short and simple, and using formatting like bullet points or holding for important information to make text easier to scan.
It is harsh to say, but we need to increasingly recognize that if your writing is largely indistinguishable from the (current) output of e.g. ChatGPT on default settings, it doesn't matter if you used ChatGPT or not, your writing is overly verbose, bad, and unpleasant to consume, and something you most certainly need to improve. I.e. your colleague needs to change his style regardless.
This sucks, but it needs to be done in education, and/or at least in areas where good writing and effective communication is considered important. Good grades need to be awarded only to writing that exceeds the quality and/or personality of a chat-bot, because, otherwise, the degree is being awarded to a person who is no more useful than a clumsy tool.
And I don't mean avoiding superficialities like the em-dash: I mean the bland over-verbosity and other systemic tells—or rather, smells—of AI slop.
> your writing is overly verbose, bad, and unpleasant to consume
Was this written by AI? Because right there we've got "three adjectives where one will do", and failing your own advice on "avoid being overly verbose"
It is up to the reader to judge whether my style is verbose, or if I could have used less adjectives here. The adjectives all in fact have different meanings, only "bad" is lazy, IMO (EDIT: and "bad" is meant to be obvious moralizing - something AI in fact almost never does).
Don't think that I don't hold myself to the same standards I am pushing here, verbosity has always been a problem for me, and AI verbosity is a good and necessary reminder for me to curb it.
> We were talking about a colleague for using ChatGPT in our WhatsApp group chat to sound smart and coming up with interesting points.
How dare they.
You're expected to infer that it wasn't working.
This is also happening to artists, people who make YouTube shorts, and similar. Everyone gets accused of being AI if the feel happens to match.
I'm sure there's some voice actor out there who can't get work because they sound too similar to the generated voices that appear in TikTok videos.
I did a video a couple years ago about a thing I did in Factorio[0] and got a couple comments who didn’t ask if I had used an AI voice, they just straight up told me that the AI voice I used was off putting. I didn’t use an AI voice, in fact I appeared on camera at the end of the video in part so that people wouldn’t have to guess, but I guess people who thought I was AI didn’t feel like watching the whole video.
I suppose I don’t mind people using AI voices if they have a thick accent or are shy about their voice, but if I’m watching a video and clock the voice as AI (usually because the tone is professional but has no expression and then the speaker mispronounces a common word or acronym) it does make me start to wonder if the script is AI. There are a lot of people churning out tutorials that seem useful at first but turn out to have no content (“draw the rest of the owl” type stuff) because they asked AI to create a tutorial for something and didn’t edit or reprompt based on the output. The video essay world is also starting to get hit pretty hard, to the point that I’m less willing than ever to watch content unless I already know the creator’s work.
[0] Shameless plug: https://youtu.be/PGiTkkMOfiw
I remember a video where the only thing that stuck with me was that the guy was working on his English skills, and used an AI voice. That's not abnormal by any stretch of the imagination nor memorable, but how he edited the AI voice made it about as natural as a normal voice.
I doubt it’s affecting his work, but my impression is that if anyone is owed money from the use of their personal likeness in AI image generators (as a source for generic figures, not as someone specifically requested by the prompt) then Pierce Brosnan is likely near the front of the queue.
I was just reading the 1897 style guide of the City News Bureau (Chicago), in the book "Hello, Sweetheart, Get Me Rewrite!". Some highlights:
- Do not confuse 'night' with 'evening'.
- This office spells it 'programme'.
- Hotels are 'kept', not 'run'.
- Dead men do not leave 'wives', but they may leave 'widows'.
- 'Very' is a word often used without discrimination. It is not difficult to express the same meaning when it is eliminated.
- The relative pronoun 'that' is used about three times superfluously to the one time that it helps the sense.
- Do not write 'this city' when you mean Chicago.
Americans have no idea what evening is. It's always night for them.
For what it's worth, LWN maintains the same attitude towards "very" today.
"Very unique" has its own problem.[1] It started appearing online around 2007, according to Google Trends. Sometimes in the description of NFTs. As an absolute adjective, "unique" should not be modified.[1] But now, "very unique" has joined the vocabulary of meaningless marketing phrases.
[1] https://www.merriam-webster.com/grammar/very-unique-and-abso...
I can only dream of writing english as well as OP. Kudos for mastering the language!
The formal part resonates, because most non-native english speaker learnt it at school, which teaches you literary english rather than day-to-day english. And this holds for most foreign languages learnt in this context: you write prose, essays, three-part prose with an introduction and a conclusion. I've got the same kind of education in france, though years of working in IT gave me a more "american" english style: straight to the point and short, with a simpler vocabulary for everyday use.
As for whether your writing is ChatGPT: it's definitely not. What those "AI bounty hunters" would miss in such an essay: there is no fluff. Yes, the sentences may use the "three points" classical method, but they don't stick out like a sore thumb - I would not have noticed should the author had not mentioned it. This does not feel like filling. Usually with AI articles, I find myself skipping more than half of each paragraph, due to the information density - just give me the prompt. This article got me reading every single word. Can we call this vibe reading?
It's the curse of writing well. ChatGPT is designed to write well, and so everyone who does that is accused of being AI.
I just saw someone today that multiple people accused of using ChatGPT, but their post was one solid block of text and had multiple grammar errors. But they used something similar to the way ChatGPT speaks, so they got accused of it and the accusers got massive upvotes.
Actually it's public info that ChatGPT was originally trained by speakers of some african business english "dialect".
https://www.theguardian.com/technology/2024/apr/16/techscape...
They said nigerian but there may be a common way English is taught in the entire area. Maybe the article author will chip in.
> ChatGPT is designed to write well
If you define well as overly verbose, avoiding anything that could be considered controversial, and generally sycophantic but bland soulless corporate speak, yes.
> They said nigerian but there may be a common way English is taught in the entire area.
Nigeria and Kenya are two very different regions with different spheres of business. I don't know, but I wouldn't expect the English to overlap that much.
There are a lot of very distinctive versions of English floating around after the British Empire, Indian newspapers are particularly delightful that way - but there is as the author says, an inherited common educational system dating back to the colonial period, which has probably created a fairly common "educated dialect" abroad, just as it has between all the local accents and dialects back in the motherland.
That's not a very good argument, because then you could say the same for American, Canada, South Africa, Australia and so on. If recency is an issue, then here's a list of colonies that got their freedom around the same time:
Cyprus, Somalia, Sierra Leone, Kuwait, Tanzania, Jamaica, Trinidad and Tobago, Uganda, Kenya, Malawi, Zambia, Malta, Gambia, Guyana, Botswana, Lesotho, Barbados, Yemen, Mauritius, Eswatini (Swaziland).
If what you're saying is right then you'd have to admit Jamaican and Barbados English are just the same as Kenyan or Nigerian... but they're not. They're radically different because they're radically different regions. Uganda and Kenya being similar is what I would expect, but not necessarily Nigeria...
>They're radically different because they're radically different regions.
They're radically different predominantly at the street level and everyday usage, but the kind of professional English of journalists, academics and writers that the author of the article was surrounded by is very recognizable.
You can tell an American from an Australian on the beach but in a journal or article in a paper of record that's much more difficult. Higher ed English with its roots in a classical British education you can find all over the globe.
That's not my experience at all. I can quite easily identify Kenyans, Australians and English from the way they write and they're all rather unique.
Go read some Kenyan news. It's very obvious.
But The Guardian could have been wrong about the country, and I'm a stupid European so I just don't know.
All we can hope is for a local to show up and explain.
ChatGPT does not “write well” unless your standard is some set of statistical distributions for vocabulary, sentence length, phrase structure, …
Writing well is about communicating ideas effectively to other humans. To be fair, throughout linguistic history it was easier to appeal to an audience’s innate sense of authority by “sounding smart”. Actually being smart in using the written word to hone the sharpness of a penetrating idea is not particularly evident in LLM’s to date.
Good writers use words to make a point. LLMs use words to make a salad.
Depends what you ask the LLM to do!
If you're using it to write in programming language, you often actually get something that runs (provided your specifications are good - or your instructions for writing the specifications are specific enough!) .
If you're asking for natural language output ... yeah... you need to watch it like a hawk by hand - sure. It'd be nice if there was some way to test-suite natural language writing.
The last time I asked it to write something in a programming language, it put together a class that seemed reasonable at first blush, but after review found it did not do what it was supposed to do.
The tests were even worse. They exercised the code, tossed the result, then essentially asserted that true was equal to true.
When I told it what was wrong and how to fix it, it instead introduced some superfluous public properties and a few new defects without correcting the original mistake.
The only code I would trust today's agents with is so simple I don't want or need an agent to write it.
Yeah, people have wildly different experiences.
I think it depends on what models you are using and what you're asking them to do, and whether that's actually inside their technical abilities. There are not always good manuals for this.
My last experience: I asked claude to code-read for me, and it dug out some really obscure bugs in old Siemens Structured Text source code .
A friend's last experience: they had an agent write an entire Christmas-themed adventure game from scratch (that ran perfectly).
But they will make the salad delicious, marvelous and intricate. It's not just a salad - it's a new way to talk like marketing copy (/s)
LLMs don't even write as well as people do. If you talk to them long enough, you'll notice they produce the same errors careless people do. Sometimes they wrongly elide the article 'a'. They occasionally mess up 'a/an' vowel agreement. The most grating thing of all is that the fully-elided 'because' (as in 'because traffic') lives on in LLM output, even though you rarely see it anymore because people rightly got the sense it was unfair for a writer to offload semantic reconstruction to the reader.
I have a confession to make: I didn't think lulcat speak was funny, even at the time.
It's pretty annoying and once you catch them doing it, you can't stop.
This may be true. I personally didn't get any hint of LLM usage from their writing. Even where they use em-dashes it's for stuff like this:
> there is - in my observational opinion - a rather dark and insidious slant to it
That feels too authentic and personal to be any of the current generation of LLMs.
ChatGPT would have used an actual em dash instead of a hyphen
Add "Always use dash instead of em dash" to the developer/system prompt, and that's never an "issue" anymore. Seems people forget LLMs are really just programmable (sometimes inaccurate) computers. Whatever you can come up with a signal, someone can come up with an instruction to remove.
Except for your poor editor who then has to manually replace your hyphens with proper em dashes. Still, if you're already disrespecting your editor enough to feed them AI slop...
My editor? I don't think it cares what I input into it, it's just a program. As long as I feed it characters it'll happily tick along as always.
The parent comment is referring to a human editor, not a text editor.
Huge assumption on their side then, isn't the context "humans writing for other humans"? Not sure how "publication editors" entered the conversation nor from where.
I was referring to a human editor, which I thought was obvious enough from context. I assumed the reply was in jest. My original comment was light-hearted, so I don't think it needs to be rigorously analysed, but plenty of humans write for other humans but still have an editor involved in the process.
They're really not programmable computers! (Bad mental model is bad.)
But yes the current commercial ones are somewhat controllable, much of the time.
Obviously not, computers are the true programmable computers. But I'd still think it's accurate to say they're like programmable computers that are sometimes inaccurate, for most intents and purposes it's a fine mental model unless you really wanna get into the weeds.
And many of us human writers would have done so, too, since we've had to learn the—not very obscure—keyboard shortcut to insert an emdash.
I would use an actual em dash if there were a keyboard key for it. On my macbook, I have an an action script set up on the touchbar for emdash and a few other unicodey glyphs, but the (virtual) buttons are like 2 inches wide each so I can't fit more than 5 or 6 across it. Sucks.
On Mac emdash is option-shift-hypen (aka shift-endash, aka capital endash)
In Menlo font (Chrome on Mac's default monospace font, used for HN comments) em-dash(—) and en-dash (–) use the same glyph, though.
Double hyphen works in many places
Depends on your definition of "well". I hate that writing style. It's the same writing style that people who want to sell you something use and it seems to be really good at tiring the reader out - or at least me.
It gives a vibe like a car salesman and I really dislike it and personally I consider it a very bad writing style for this very reason.
I do very much prefer LLMs that don't appear to be trained on such data or try to word questions a lot more to have more sane writing styles.
That being said it also reminds me of journalistic articles that feel like the person just tried to reach some quota using up a lot of grand words to say nothing. In my country of residence the biggest medium (a public one) has certain sections that are written exactly like that. Luckily these are labeled. It's the section that is a bit more general, not just news and a bit more "artsy" and I know that their content is largely meaningless and untrue. Usually it's enough to click on the source link or find the source yourself to see it says something completely different. Or it's a topic that one knows about. So there even are multiple layers to being "like LLMs".
The fact that people are taught to write that way outside of marketing or something surprises me.
That being said, this is just my general genuine dislike of this writing style. How an LLM writes is up to a lot of things, also how you engage with it. To some degree they copy your own style, because of how they work. But for generic things there is always that "marketing talk" which I always assumed is simply because the internet/social media is littered with ads.
Are Kenyans really taught to write that way?
Are Kenyans really taught to write that way?
I’m highly skeptical. At one point the author tries to argue this local pedagogy is downstream of “The Queen’s English” & British imperial tradition, but modern LLM-speak is a couple orders of magnitude closer in the vector space to LinkedIn clout-chasing than anything from that world.
Yes they are, or rather, we were when I was in primary school. My essays (we called them composition) were filled with these these red check marks for every esoteric word, proverb, metaphor or simile you used. The more you had the higher you'd score. So I did my homework with a dictionary open. I remember writing some document at work in the US and everyone commenting on how Queen's English it was. This was before ChatGPT. I know know it was all silly, and I've spent a bunch of time learning to write simply. But then I've listen to too many tech podcasts, and now I find silicon valley tech-speak creeping in, and I hate it. The one that I hear everywhere now that I swear not to ever use is let's double-click on that point. Just why?
Sure, I believe that completely. But that's not how ChatGPT writes!
Here are some random examples from one of the (at least) half-dozen LLM-co-written posts that rose high on the front page over the weekend:
https://blog.canoozie.net/disks-lie-building-a-wal-that-actu...
You write a record to disk before applying it to your in-memory state. If you crash, you replay the log and recover. Done. Except your disk is lying to you.
This is why people who've lost data in production are paranoid about durability. And rightfully so.
Why this matters: Hardware bit flips happen. Disk firmware corrupts data. Memory busses misbehave. And here's the kicker: None of these trigger an error flag.
Together, they mean: "I know this is slower. I also know I actually care about durability."
This creates an ordering guarantee without context switches. Both writes complete before we return control to the application. No race conditions. No reordering.
... I only got about halfway through. This is just phrasing, forget about the clickbaity noun-phrase subheads or random boldface.
None of these are representative (I hope!) of the kind of "sophisticated" writing meant to reinforce class distinctions or whatever. It's just blech LinkedIn-speak.
And good students are getting in trouble (meaning "have to explain themselves") to lousy teachers just because they write well, articulate ideas and can summarize information from documents where other regular people would make mistakes.
> they used something similar to the way ChatGPT speaks, so they got accused of it and the accusers got massive upvotes
Outrage mills mill outrage. If it wasn't this, it would be something else. The fact that the charge resonated is notable. But the fact that it exists is not.
ChatGPT writes a particular dialect of good writing. Always insisting on cliffhangers towards the summary, or "strong enumerations", like "the candidate turned out to be a bot. Using ChatGPT. Every. Single. Time." And so on.
"Every. Single. Time." has been a staple of American online humor for at least a decade. Commonly used, hence commonly used by ChatGPT.
I saw this described as LLMs writing "punched up" paragraphs, and every paragraph must be maximally impacting. Where a human would acknowledge some paragraphs are simply filler, a way to reach some point, to "default" LLMs every paragraph must have maximum effect, like a mic drop.
It's the content mill blogspam voice. The machine generated slop looks a lot like the artisan hand crafted slop.
This reminds me of Idiocracy: "Ah, you talk like a fag, and your shit's all retarded" as a response to a normal speech.
Ironically OpenAI used Kenyan workers[1] to train its AI and now we've come to the point where Kenyans are being excluded because they sound too much like the AI that they helped train.
[1] https://time.com/6247678/openai-chatgpt-kenya-workers/
I actually think that's a great endorsement of Kenyan education. I don't deal with English-speaking African countries that often (I'm Portuguese, so naturally we have ties to other bits of the continent), but I've often been impressed by how well they communicate regardless of the profession they're in--I don't mean that as a bias, but rather as it befitting the kind of conversation you'd have with an English major in the UK (to which I have a lot of exposure).
Perhaps the US-centric "optimization" of English is to blame here, since it is so obvious in regular US media we all consume across the planet, and is likely the contrasting style.
It's not ironic
I think it is. The irony is that the people you hired to help make your machine seem human are seen as mechanical because of their distinct and uniformly sophisticated tone. Thus we have a situation that’s contrary to expectations.
Also Kenyan, I once recently spent 10min explaining a technical topic via chat, and the response I got was "was this GPT?". I took a few minutes then just linked an article of how underpaid Kenyans trained ChatGPT for OpenAI [1]
1: https://time.com/6247678/openai-chatgpt-kenya-workers/
> I don't write like ChatGPT. ChatGPT, in its strange, disembodied, globally-sourced way, writes like me.
We will all soon write and talk like ChatGPT. Kids growing up asking ChatGPT for homework help, people use it for therapy, to resumes, for CVs, for their imaginary romantic "friends", asking every day questions from the search engine they'll get some LLM response. After some time you'll find yourself chatting with your relative or a coworker over coffee and instead of hearing, "lol, Jim, that's bullshit" you'll hear something like "you're absolutely right, here let me show you a bulleted list why this is the case...". Even more scarier, you'll soon hear yourself say that to someone, as well.
(star-eyes emoji) You are absolutely correct, Jim!
(check-mark emoji) Add more emoji — humans love them! (red x emoji) Avoid negative words like "bullshit" and "scarier."
(thumbs-up emoji) Before long you'll get past the human feedback of reinforcement learning! (smiley-face)
Ironically, mistakes and idiosyncrasies are becoming a sign of authenticity and trustworthiness, while polish and quality signal the opposite.
Earlier today I stumbled upon a blog post that started with a sentence that was obviously written by someone with a slavic background (most writers from other language families create certain grammatical patterns when writing in another language, e.g. German is also quite typical). My first thought was "great, this is most likely not written by a LLM".
It's an age-old cycle in media. There have been innumerable waves of more gritty aesthetic trends when things became too polished or inane: jazz, rock, punk, rap, hippies, goths, hipsters, 70s cinema, HBO golden-age, YouTube, blogging, early social media, even MAGA...
Authenticity, wether it is sincere or not, can become an incredibly powerful force now and then. Regardless of AI, the communication style in tech, and overall, was bound to go back to basics after the hacker culture of the post-dotcom era morphed, in the 2010s, into the corporatism they were fighting to begin with, yet again.
Very good point, also in classic art history, you often had a sequence of a period that perfected a certain style until it became formalistic, and then a subsequent one that broke off with the previous style, like Renaissance->Mannerism, Baroque->Rococo, Classicism,Realism,Photography->Impressionism, etc.
Maybe for writing, but in digital art circles if anyone notices a mistake in your lines or perspective or any kind of technical error you will get the anti-AI cancel mob after you even if you didn’t use generative AI at all.
I would not want to be an artist in the current environment, it’s total chaos.
I'm an artist in the current environment, it's not total chaos. Ignore what others are doing, do what you want with the tools you have available, and you'll be fine. There are huge echo-chambers on the internet, but once you get out in the real world, things are not as people on the internet paints it out to be.
> artist
Social media artists, gallery artists and artists in the industry (I mean people who work for big game/film studios, not industrial designers) are very different groups. Social media artists are having it the hardest.
with a sentence that was obviously written by someone with a slavic background
Omitting articles? To me, that has always signaled "this will be an interesting and enlightening read, although terse and in need of careful thought." I've found sites from that part of the Internet to be very useful for highly technical and obscure topics.
To an extent this has always been the case (this kid has clearly made a strong attempt at following some quite basic instructions, versus this kid's answer is - perhaps literally - textbook).
But yeah, I definitely find mild grammatical quirks expected from English as a foreign language speakers a positive these days, because the writing appears to reflects their actual thoughts and actual fluency.
until you ask it write like this, because why use many word when few do trick?
AI is not only replacing us, it's forcing us to self-dumb down too!
Always interesting (in an informative way) to see people "defending" em-dashes from my personal perspective. Before you get mad, let me explain: before ChatGPT, I only ever saw em-dashes when MS Word would sometimes turn a dash into a "longer dash" as I always thought of it. I have NEVER typed an em-dash, and I don't know how to do it on Windows or Android. I actually remember having issues with running a program that had em-dashes where I needed to subtract numbers and got errors, probably from younger me writing code in something other than an IDE. Em-dashes always seem very out of place to me.
Some things I've learned/realized from this thread:
1. You can make an em-dash on Macs using -- or a keyboard shortcut
2. On Windows you can do something like Alt + 0151 which shows why I have never done it on purpose... (my first ever —)
3. Other people might have em-dashes on their keyboard?
I still think it's a relatively good marker for ChatGPT-generated-text iff you are looking at text that probably doesn't apply to the above situations (give me more if you think of them), but I will keep in mind in the future that it's not a guarantee and that people do not have the exact same computer setup as me. Always good to remember that. I still do the double space after the end of a sentence after all.
Well, (some) people on HN definitely used them before ChatGPT. https://www.gally.net/miscellaneous/hn-em-dash-user-leaderbo...
(And as #9 on the leaderboard, I feel the need to defend myself!)
Maybe I'm weird, but one of the first things I've always done when setting up emacs is to enable Typo mode (or Typopunct) for writing modes, which handles typing en and em dashes and "smart" quotation marks in a fairly natural way.
I actually checked HN's comment data corpus to see if em dash usage rose after AI adoption became more widespread. I was kind of shocked to see that it did not.
Its overuse is definitely a marker of either AI or a poorly written body of text. In my opinion, if you have to rely on excessive parentheticals, then you are usually off restructuring your sentences to flow more clearly.
I actually got punked during a demo because I wrote some terminal commands and stored them in the macOS notepad and didnt notice it had changed -- to —.
When I copy and pasted them in it failed obviously so... yeah. If you have terminal commands that use `--` don't copy+paste them out of notepad.
Shift+Win/Option+-. And holding - gives you en/em dash on iOS and Android. Personally I love using em dashes so this whole AI thing is a real disaster for me.
Just a reminder that our experience does not necessarily invalidate someone else's experience.
Eg, I was typing Alt-0151 and Alt-0150 (en-dash) on the reg in my middle school and high school essays along with in AIM. While some of my classmates were probably typing double hyphens, my group of friends were using the keyboard shortcuts, so I am now learning from this "detect an LLM" faze that there's a vocal group of people who do not share this experience or perspective of human communication. And that having a mother who worked in technical publishing who insisted I use the correct punctuation rather than two hyphens was not part of everyone's childhood.
I read about 4 paragraphs of the blog post, it does not at all read like it was written by ChatGPT!
Some people are perhaps overly focussed on superficial things like em-dashes. The real tells for ChatGPT writing are more subtle -- a tendency towards hyperboly (it's not A, it's [florid restatment of essentially A] B!), a certain kind of rhythym, and frequently a kind of hard to describe "emptiness" of claims.
(LLMs can write in mang styles, but this is the sort of "kid filling out the essay word count" style you get in chatgpt etc by default.)
It does not, but to many, many people who cannot tell the difference it does. Simply because it's well-written somewhat-formal-register English and not "internet speech" or similar casual register. As you probably know, there are many these days who take the mere use of em or en dashes as a reliable sign of LLM writing.
Hey bro! This is the real English bro! No way we can write like that bro! What? - and ;? The words like "furthermore" or "moreever"? All my homies nver use the words like that bro! Look at you. You're using newline! You're using ChatGPT, right bro?
Given the eloquently natural words in this post, I conclude you must be this thread's prompt engineer! Well done, my fellow Netizen. Reading your words was like smelling a rosebud in spring, just after the heavy snow fell.
Now, please, divulge your secret--your verbal nectar, if you wish--so that I too can flower in your tounge!
I'm not sure I've read any of Marcus' previous writing, but there's no way that essay could have been written by an AI. It's personal and has a structure that follows human thought rather than a prompt.
For sure he describes an education in English that seems misguided and showy. And I get the context - if you don't show off in your English, you'll never aspire to the status of an Englishman. But doggedly sticking to anyone's "rules of good writing" never results in good writing. And I don't think that's what the author is doing, if only because he is writing about the limitations of what he was taught!
So idk maybe he does write like ChatGPT in other contexts? But not on this evidence.
I have seen people use "you're using AI" as a lazy dismissal of someone else's writing, for whatever reasons. That usually tells you more about the person saying it than the writing though.
I see people claiming real videos are AI, or even real photos. You can really tell it's not when there's 17 other videos from other angles. Maybe someday AI will get good at that level of faking a video, but at the time being, it is much harder to pull off.
In russian there is a saying that translates "try to prove that you are not a camel" which describes the impossibility of proving what is obviously not true to an unwilling and/or obtuse party.
According to russian language wikipedia (https://ru.wikipedia.org/wiki/%D0%94%D0%BE%D0%BA%D0%B0%D0%B7...) the original tale go out to famous Persian poet Rumi from XII century, which just makes me tickled pink about how awesome language is.
I always thought the whole argument was about explicitly using em dash and / or en dash. Aka — and –.
Because while people OBVIOUSLY use dashes in writing, humans usually fell back on using the (technically incorrect) hyphen aka the "minus symbol" - because thats whats available on the keyboards and basically no one will care.
Seems like, in the biggest game of telephone called the internet, this has devolved into "using any form of dash = AI".
Great.
The funniest thing I see are people who are harking "Eww, you used AI for this and it's bad because of that, I can tell because I used this other AI service who said what you wrote was 90% of AI", completely failing to grasp the irony.
Yeah, the joys of mass ignorance.
- Barely literate native English speakers not comprehending even minimally sophisticated grammatical constructs.
- Windows-centric people not understanding that you can trivially type em-dash (well, en-dash, but people don’t understand the difference either) on Mac by typing - twice.
Recently, many people do use the em dash. One big reason is that iOS and I think macOS auto converts a double - into an em dash.
> and basically no one will care
Wow, you really do under/over estimate some of us :)
Fair. I was probably just projecting. I cant even figure out when to use a comma in my native language. So caring about which type of hyphen was used feels like overly sophisticated to me - because I dont care myself.
Ah, no, I was only joking. I may be a grammar pedant, but I can also do self-deprecation.
> You didn’t just ‘walk’; you ‘strode purposefully’, ‘trudged wearily’, or ‘ambled nonchalantly’.
‘Striding’ is ‘purposeful’; ‘trudging’ expresses ‘weariness’; ‘ambling’ implies ‘nonchalance’.
Good verb choice reduces adverb dependence.
It is a shame that the author has to change to keep up, and I feel their pain but .. it’s also the price of progress. We all do things to keep up when change comes for our work and skill sets.
LLMs - like all tools - reduce redundant & repetitive work. In the case of LLMs it’s now easy to generate cookie cutter prose. Which raises the bar for truly saying something original. To say something original now, you must also put in the work to say it in an original way. In particular by cutting words and rephrasing even more aggressively, which saves your reader time and can take their thinking in new directions.
Change is a constant, and good changes tend to gain mass adoption. Our ancestors survived because they adapted.
I had a similar experience recently during code review. I was told to remove extra comments produced by Cursor. I was like, “I didn’t use Cursor for any of this PR…”
I also love and use em-dashes regularly. ChatGPT writes like me.
It's an arms race between human writers and AI. Writers want to sound less like AI and AI wants to sound more like writers, so no indicator is reliable for long. Today typos indicate a real writer, so tomorrow LLMs will inject them where appropriate. Yesterday em dashes indicated LLM, so now LLMs use them less.
Beyond these surface level tells though, anyone who's read a lot of both AI-unassisted human writing as well as AI output should be able to pick up on the large amount of subtler cues that are present partly because they're harder to describe (so it's harder to RLHF LLMs in the human direction).
But even today when it's not too hard to sniff out AI writing, it's quite scary to me how bad many (most?) people's chatbot detection senses are, as indicated by this article. Thinking that human writing is LLM is a false positive which is bad but not catastrophic, but the opposite seems much worse. The long term social impact, being "post-truth", seems poised to be what people have been raving / warning about for years w.r.t other tech like the internet.
Today feels like the equivalent of WW1 for information warfare, society has been caught with its pants down by the speed of innovation.
> society has been caught with its pants down by the speed of innovation.
Or rather by the slowness of regulation and enforcement in the face of blatant copyright violation.
We've seen this before, for example with YouTube, which became the go-to place for videos by allowing copyrighted material to be uploaded and hosted en masse, and then a company that was already a search engine monopoly was somehow allowed to acquire YouTube, thereby extending and reinforcing Google's monopolization of the web.
Innovation has always been faster when copyright is lax. The US was copying British and other European inventions during the industrial age left and right, and their economy took off because of it.
This author writes in ESL better than 99% of the people I've worked with in an English-native country, including myself. It's fascinating to read just how much more emphasis good-quality written English seems to have in Kenya than it does here in Australia (at least in the public education system where I have experience). I suppose that it's understandable, given that it gates access to higher-level education opportunities.
I don't really understand the aversion some people have to the use of LLMs to generate or refine written communication. It seems trigger the "that's cheating!" outrage impulse.
I don't think the author mentioned that English is their second language. English is an official language of Kenya, and there's a reasonable chance it's the author's home language.
AI / LLMs, including ChatGPT, can already be made to sound (almost) any way you want, just by telling it to. The usual tells that something was written or created by AI are changing monthly.
Just recently I was amazed with how good text produced by Gemini 3 Pro in Thinking mode is. It feels like a big improvement, again.
But we also have to honest and accept that nowadays using a certain kind of vocabulary or paragraph structure will make people think that that text was written by AI.
Actors have known this for decades: self-expression isn’t only a stage problem. It’s a life problem. Most people fail to express themselves on an hourly basis. Being good at expressing yourself is unnatural. Having clarity of what “yourself” even is is unnatural. The truth is that we’re all making comments, jokes, deciding what’s important and what not using old programming in our brains… programming that was given to us by our childhood and our education. Very few people can consistently have the luxury of being/ability to be creative with that old programming, and even those that can often have to plan ahead of time/rigidly control the environment in order to achieve a creative result.
The exact same problem exists with writing. In fact, this problem seems to exist across all fields: science, for example, is filled with people who have never done a groundbreaking study, presented a new idea, or solved an unsolved problem. These people and their jobs are so common that the education system orients itself to teach to them rather than anyone else. In the same way, an education in literature focused on the more likely traits you’ll need to get a job: hitting deadlines, following the expected story structure, etc etc.
Having confined ourselves to a tiny little box, can we really be surprised that we’re so easy to imitate?
Funny how sci-fi always envisioned AI to speak in a rigid, hyper-rational terseness, whereas reality gave us AI which inherited the worst linguistic vices of "human" voices.
You call writing in a structured fashion with formal words the "worst linguistic vices"
The worst vices are the superfluous faux-eloquence that meanders without meaning. Employing linguistic devices for the sake of utilizing them without managing to actually make a point with its usage.
I was trying to figure out why my SD card wasn't mounting and asked ChatGPT. It said:
> Your kernel is actually being very polite here. It sees the USB reader, shakes its hand, reads its name tag… and then nothing further happens. That tells us something important. Let’s walk this like a methodical gremlin.
It's so sickly sweet. I hate it.
Some other quotes:
> Let’s sketch a plan that treats your precious network bandwidth like a fragile desert flower and leans on ZFS to become your staging area.
> But before that, a quick philosophical aside: ZFS is a magnificent beast, but it is also picky.
> Ending thought: the database itself is probably tiny compared to your ebooks, and yet the logging machinery went full dragon-hoard. Once you tame binlogs, Booklore should stop trying to cosplay as a backup solution.
> Nice, progress! Login working is half the battle; now we just have to convince the CSS goblins to show up.
> Hyprland on Manjaro is a bit like running a spaceship engine in a treehouse: entirely possible, but the defaults are not tailored for you, so you have to wire a few things yourself.
> The universe has gifted you one of those delightfully cryptic systemd messages: “Failed to enable… already exists.” Despite the ominous tone, this is usually systemd’s way of saying: “Friend, the thing you’re trying to enable is already enabled.”
Did you not put some weird thing in your prompt ? That's not the style of writing I have in my ChatGPT, I run without memory and with default prompt. Yours try to make a metaphore at every single response.
You can check both in ChatGPT settings.
These are cherry picked. Mostly the first and last sentence look like this.
I just checked settings, apparently I had it set to "nerdy," that might be why. I've just changed it to "efficient," hopefully that'll help.
Probably because we're discussing the chatbot form of AI rather than a more general one.
That's because there were only so many lines of Spock's dialogue to train an LLM on, they needed more and so trained them on reddit comments instead.
Bang on. The self proclaimed detectives have never had to take TOEFL where you'll get marks deducted for not using connectors like furthermore.
Goodness, I forgot about TOEFL. That might indeed shape a lot of your early vocabulary choices if you need to get an English certificate (which I suppose would happen during college years, which is also when most of your personal writing style gels together).
I firmly believe that the heuristics that teachers/lecturers/instructors worldwide use to avoid engaging with reams of mundane text have been successfully by LLMs, and that's why they were so hostile to them initially.
They have to actually read material, and not just use the structure as a proxy for ability.
Pangram, the best AI-detector I know of, flagged this as 100% AI generated.
That's just sad. I really feel for this author.
If you think about societies still in English colonial hangover and ChatGPT you might find that they have similar reasons to speak the way they speak.
Both aim at using an English that is safe, controlled and policed for fear of negative evaluation.
I use semi-colons and em-dashes liberally too. But I tend to do a second pass to avoid redundancy.
e.g. > [...] and there is - in my observational opinion - a rather dark and insidious slant to it
Let's leave it at "insidious" and "in my opinion". Or drop "in my opinion" entirely, since it goes without saying.
Just take one dip and end it.
(https://www.youtube.com/watch?v=RfprRZQxWps)
If you are "average", you will sound like an AI, because an AI is the average of its training data. I don't think it's only Kenyans; I've seen the same distinctive "dialect" from many others.
The "voice" of chatbots comes from the stuff after the main training, which is very different to the average human voice. Even the main training would give you something like "average on the internet" voice, which is quite different to the average human voice.
The internet been the same for a long time, it's just the wording that changed. As someone who apparently thinks differently, the amount of time people just end up saying "Well, you're just a troll, no one actually believes something like that, so whatever" since I started frequenting the internet in the early 2000s is the same as always. But some people try to be trendy and accuse you of using AI for writing the replies instead, but it's the same sentiment.
Besides, of course what people write will sound as LLMs, since LLMs are trained on what we've been writing on the internet... For us who've been lucky and written a lot and are more represented in the dataset, the writings of LLMs will be closer to how we already wrote, but then of course we get the blame for sounding like LLMs, because apparently people don't understand that LLMs were trained on texts written by humans...
I feel this a bit, since I'm a voracious reader and a constant writer across a few languages (but mostly English), which over the decades has led to my converging on a certain (if imperfect) degree of polish. Plus my multiple concurrent and often fragmented simultaneous trains of thought while writing lead me to use parentheticals very often while drafting, which then means I often need go back and re-introduce structure.
And guess what, when you revise something to be more structured and you do it in one sitting, your writing style naturally gravitates towards the stuff LLMs tend to churn out, even if with less bullet points and em dashes (which, incidentally, iOS/macOS adds for me automatically even if I am a double-dash person).
What gets me these days is sort of this structure.
X isnt just Y its a <description> Z!
This post doesn't read anything like ChatGPT. Correct grammar does not indicate ChatGPT. Em-dashes don't indicate ChatGPT. Assessing whether something was generated using an LLM requires multiple signals, you can't simply decry a piece of text as AI-generated because you noticed an uncommon character.
Unfortunately I think posts like this only seem to detract from valid criticisms. There is an actual ongoing epidemic of AI-generated content on the internet, and it is perfectly valid for people to be upset about this. I don't use the internet to be fed an endless stream of zero-effort slop that will make me feel good. I want real content produced by real people; yet posts like OP only serve to muddy the waters when it comes to these critiques. They latch onto opinions of random internet bottom-feeders (a dash now indicates ChatGPT? Seriously?), and try to minimise the broader skepticism against AI content.
I wonder whether people like the Author will regret their stance once sufficient amount of people are indoctrinated and their content becomes irrelevant. Why would they read anything you have to say if the magic writing machine can keep shitting out content tailored for them 24/7?
I’m having a similar problem. Spent way too much time on the internet starting in my preteens and it shaped the way I write - which not surprisingly - is a similar way to how an AI - trained on the online data - writes
Excellent article. Insightful observation, expressed and written well. Just an opinion from a Canadian borne and American raised native English speaker.
On social media I've been accused of being AI twice now :-). I suspect it is a vocabulary thing but still it is always amusing.
I wouldn't usually use the 'non-native speaker argument', but thank you! Just yesterday I was accused of sounding like AI - https://news.ycombinator.com/item?id=46262777 - my default mode is that I oscillate between sounding too boring/technical, or when trying to do my best, sounding like AI
Correlated but kinda off topic: I don't mind the style so much, I mind the verbosity. The amount of words spit out effortless by the writer which then need to be comprehended and filtered by every reader.
Seeing a project basically wrapping 100 lines of code with a novel length README ala 'emoticon how does it compare to.. emoticon'-bla bla really puts me off.
That's a hallmark of Claude. I stopped using Claude for documentation because it was overly... JavaScripty in feel (all the stuff it churned out felt like JavaScript framework docs of the 2010s, and I bet it would have added Neon Cat if it knew how).
In comparison, I can sort of confidently ask GPT-5.1/2 to say "revise this but be terse and concise about it" and arrive at something that is more structured that what I input but preserves most of my writing style and doesn't bore the reader.
We shouldn't need to have people bearing false witness. Anyone who uses AI tools to produce published works should offer a clear disclaimer to their audience. I share the same concerns as the author: "Will my written work be used to say that I plagiarize off ChatGPT?"
All the toil of word-smithing to receive such an ugly reward, convincing new readers that you are lazy. What a world we live in.
The fact that everyone is now constantly forced to use (oftentimes faulty) personal heuristics to determine whether or not they read slop is the real problem here.
AI companies and some of their product users relentlessly exploit the communication systems we've painstakingly built up since 1993. We (both readers and writers) shouldn't be required to individually adapt to this exploitation. We should simply stop it.
And yes, I believe that the notion this exploitation is unstoppable and inevitable is just crude propaganda. This isn't all that different from the emergence of email spam. One way or the other this will eventually be resolved. What I don't know is whether this will be resolved in a way that actually benefits our society as a whole.
> fact that everyone is now constantly forced to use (oftentimes faulty) personal heuristics to determine whether or not they read slop is the real problem here
It would be ironic and terrific if AI causes ordinary Americans to devote more time to evaluting their sources.
It could be also because, we foreigners learn to write English prose through our reading comprehension, not via our listening circuitry. The text probably feels "normal" to me when I read it back to myself, but there's no "proper" feedback loop from the native speakers - I have zero idea how my written shit sounds to a native ear when they try to read it. I still do agree though, it feels so friggin' annoying these days to have to deliberately butcher some words and make sure there's a typo somehwere in your text, just to convince people that tis indeed a crap straight from my "head to the paper", not a slop.
Even more so, I think most of the curated data for the fine tuning phase is hand crafted from people from countries like Kenya if I recall.
This must be infuriating:
> You spend a lifetime mastering a language, adhering to its formal rules with greater diligence than most native speakers, and for this, a machine built an ocean away calls you a fake.
This is :
> humanity is now defined by the presence of casual errors, American-centric colloquialisms, and a certain informal, conversational rhythm
And once you start noticing the 'threes', it's fun also.
I mean, "to err is human" was written in the 1700s, by the enlightenment era author the essay writer is presumably reading.
Humanity has always been about errors.
If you used a calculator to do a calculation, would they say the answer looks like created by calculator and not done by-hand?
I think the only solution to this is, people should simply not question AI usage. Pretence is everywhere. Face makeup, dress, the way you speak, your forced smile...
> Human touch. Human touch. I’ll give you human touch, you—
> TECHNICAL DIFFICULTIES PLEASE STAND BY
This actually made me pee myself out loud!
I don't mind the "normal" text so much, where you aren't sure if it was written by an AI or not. What's really getting annoying is the flood of bullet points and emoji that is flooding LinkedIn in particular. Super obnoxious!
Love this, everything about this - I still teach the foundation, 3 columns, roof, of the persuasive essay - except one bit:
Perplexity gauges how predictable a text is. If I start a sentence, "The cat sat on the...", your brain, and the AI, will predict the word "floor."
No. No no no. The next word is "mat"!
rat. And its claws dug in its back.
How do you like that, Mr. Rat
Thought the Cat.
ha ha - I had the same thought!
Well, his writing style is too good. The sentences flow too beautifully, he uses rich vocabulary and styling. It's unusual to see that style of writing online. I definitely don't poses that power.
I don't know the author of this article and so I don't know whether I should feel good or bad about this. LLMs produce better writing than most people can and so when someone writes this eloquently, then most people will assume that it's being produced by LLM. The ride in the closed horse carriage was so comfortable it felt like being in a car and so people assumed it was a car. Is that good? Is that bad?
Also note that LLMs are now much more than just "one ML model to predict the next character" - LLMs are now large systems with many iterations, many calls to other systems, databases, etc.
> LLMs produce better writing than most people can and so when someone writes this eloquently, then most people will assume that it's being produced by LLM.
I really don’t think that is what most normal people assume… And while LLMs can definitely produce more grammatically accurate prose with probably a wider vocabulary than the average person, that doesn’t necessarily mean it’s good writing…
I meant "good" in the formatting, grammar, vocabulary sense. I'm not arguing that LLMs are "good" in writing amazing prose.
I mean look at two of us - I have typos, I use half broken english, I'm not good in doing noun articles, my vocabulary is limited, I don't connect sentences well, you end sentences with "..." and then you start sentence with "And", etc. I very much believe you are a real person.
Systemic discrimination, happens all the time. I am blind. I regularily fail the "tell computers and humans apart" test. You imagine, that feels very much like the dehumanisation it is. Big tech couldn't care less. After all, they need to protect themselves against spammers. Much like the guy who was on the HN frontpage just a few days ago, arguing that he is now trashing accessibility because he doesn't want to be web scraped. If you raise these issues with devs, all you get it pushback, no understanding at all. Thats the way it is. If you are amongst a minority small enough and without a rainbow coloured flag, you end up being ignored, stepped over, and pushed aside. If you are lucky. If you are unlucky, and you raise your voice, you will be critizied for pointing out the obvious.
I agree anti-bot vigilantes as well as corporate anti-ddos middle-wares have had a detrimental impact on accessibility. I'm afraid they consider your use case as acceptable collateral damage if they consider it at all.
I know... Its depressing.
> arguing that he is now trashing accessibility because he doesn't want to be web scraped
Interesting, because he failed me too just because I use Firefox. Have you been told about the article or it actually worked with your screen reader software?
I have to admit I only read the heading. I didn't want to read the article, that would have ruined my day.
He messed with the glyph indexes in a customized font so the text is gibberish if you look just at the code points but displays as english.
That would probably mess up any screen reader, but it also didn't work on a regular Firefox :)
wasn't that the article about the obfuscation of kindle ebooks?
https://news.ycombinator.com/item?id=46264955
No, don't think so. To compensate, I probably missed the article about the obfuscation of kindle ebooks...
https://news.ycombinator.com/item?id=45610226
Hmm 2 months ago. Now I wonder if the link you posted inspired the link I posted...
What was the "dead giveaway" referred to in the pasted tweet? Was it the dash, that people assume for some reason regular folks never use? Or was it something more interesting?
the initial rlhf training evaluation was done by kenyans specifically
Thank you for writing this. I too was a heavy user of the em-dash until ChatGPT came along. Though my solution has been to eschew the em-dash or at least replace with triple hyphens.
It's pretty rude to "accuse" someone of using AI. Would you yell "Dictionary!", "Grammarly!", "Reference manual!", "Newspaper quote!" at them? Maybe "Harvard!" or "Tutored!" ? You don't know who they are or what their life is like. Maybe they're blind and using it as an assisted device. Maybe their hand is injured and they use it to output information faster. Maybe they're old, infirm, a non-native English speaker, etc. Maybe they're just a regular person who feels insecure writing, and wants to use new technology to give them the confidence to write/comment more. Or, maybe they just talk like that.
Let's say you happen to be lucky, don't accuse someone unfairly, and they are using ChatGPT to write what they said. Who cares?! What is it you're doing by "calling them out" ? Winning internet points? Feeling superior? Fixing the world?
> Let's say you happen to be lucky, don't accuse someone unfairly, and they are using ChatGPT to write what they said. Who cares?!
People who want to read thoughts of other people and not meaningless slop.
ChatGPT definitely writes like it's trainers (like Kenya).
Kenya writes like the British taught before they left, and necessarily they didn't speak or write how they did.
TIL Kinyarwanda is the national language of Rwanda, not Kenya
It actually really bothers me that somehow the long dash has become such a "giveaway" that now I have to consider how I write. I actually used the long dash a lot in my normal writing, but now that everyone considers it some sort of "giveaway," it's like a tic that I can't help but notice.
It feels very natural to me. But if everyone and their mother considers it a "giveaway", I'd be a fool not to consider it. * sigh *
> For my generation, and the ones that followed, the English Composition paper - and its Kiswahili equivalent, Insha - was not just a test; it was a rite of passage.
OK but come ON, that has to have been deliberate!
In addition to the things chatbots have made clichés, the author actually has some "tells" which identify him as human more strongly. Content is one thing. But he also has things (such as small explanations and asides in parentheses, like this) which I don't think I've EVER seen an instruction-tuned chatbot do. I know I do it myself, but I'm aware it's a stylistic wart.
This resonates with me. LLM output in Spanish also has the tendency to "write like me", as in the linked article.
On that regard, I have an anecdote not from me, but from a student of mine.
One of the hats I wear is that of a seminary professor. I had a student who is now a young pastor, a very bright dude who is well read and is an articulate writer.
"It is a truth universally acknowledged" (with apologies to Jane Austen) that theological polemics can sometimes be ugly. Well, I don't have time for that, but my student had the impetus (and naiveté) of youth, and he stepped into several ones during these years. He made Facebook posts which were authentic essays, well argued, with balanced prose which got better as the years passed by, and treating opponents graciously while firmly standing his own ground. He did so while he was a seminary student, and also after graduation. He would argue a point very well.
Fast forward to 2025. The guy still has time for some Internet theological flamewars. In the latest one, he made (as usual) a well argued, long-form Facebook post, defending his viewpoint on some theological issue against people who have opposite beliefs on that particular question. One of those opponents, a particularly nasty fellow, retorted him with something like "you are cheating, you're just pasting some ChatGPT answer!", and pasted a screenshot of some AI detection tool that said that my student's writing was something like "70% AI Positive". Some other people pointed out that the opponent's writing also seemed like AI, and this opponent admitted that he used AI to "enrich" some of his writing.
And this is infuriating. If that particular opponent had bothered himself to check my student's profile, he would have seen that same kind of "AI writing" going on back to at least 2018, when ChatGPT and the likes were just a speck in Sam Altman's eye. That's just the way my student writes, and he does in this way because the guy actually reads books, he's a bonafide theology nerd. Any resemblance of his writing to a LLM output is coincidence.
In my particular case, this resonated with me because as I said, I also tend to write in a way that would resemble LLM output, with certain ways to structure paragraphs, liberal use of ordered and unordered lists, etc. Again, this is infuriating. First because people tend to assume one is unable to write at a certain level without cheating with AI; and second, because now everybody and their cousin can mimic something that took many of us years to master and believe they no longer need to do the hard work of learning to express themselves on an even remotely articulate way. Oh well, welcome to this brave new world...
This essay is effin' brilliant, and beautifully written.
I'm not Kenyan, but I was raised in a Canadian family of academics, where mastering thoughtful – but slightly archaic – writing was expected of me. I grew up surrounded by books that would now be training material, and who's prose would likely now be flagged as ChatGPT.
Just another reason to hate all this shit.
Looking forward to the deliberately abstruse and illogical essays of the future. Everyone will have to write like a second-rate French philosopher.
"Rewrite this email paragraph in the style of a corporate ToS statement. Do NOT expose my orders and their implicit acceptance of them by the recipient pending a 24 hr deadline anywhere before page 18."
Honestly, people assuming I'm using ChatGPT to communicate with them and liberally using that suspicion as a filter sounds like a great meta-filter.
""The cat sat on the...", your brain, and the AI, will predict the word "floor.""
The models mostly say "mat".
It is highly inappropriate to accuse anyone with the claim that their writing is AI generated. This often is used as an excuse to unfairly discredit the content of the message. Whether the content is or isn't AI generated can't be determined with any confidence, and even if it is, it is improper to ignore the message. If you're going to criticize a message, do so on the basis of its actual content, not its alleged authorship.
I can't wait until we reach the point of AI adoption where genuine content is suspicious.
Wanna submit a proof in a criminal case? Better be ready to debunk whether this was made with AI.
AI is going to fuck everything up for absolutely no reason other than profit and greed and I can't fucking wait
It's going to make accountability very, very difficult. We were nearly at the point in politics anyway, where people could just claim evidence was fake and get away with it. Now, it's an easy get-out. I am fully expecting that, if any particularly incriminating photos were to appear, say of powerful people engaging in activities with Jeffrey Epstein, that they will simply dismiss them as "fake news AI".
BS. ChatGPT writes in the sterile and boring manner of the average graduate of business, marketing or journalism: it is dull, safe, somewhat pompous but professional, the ideal style for corporate communication.
Basically, for two reasons:
1) A giant portion of all internet text was written by those same folks. 2) Those folks are exactly the people anyone would hire to RLHF the models to have a safe, commercially desirable output style.
I am pretty convinced the models could be more fluent, spontaneous and original, but then it could jeopardize the models' adoption in the corporate world, so, I think the labs intentionally fine-tuned this style to death.
The tell isn't the emdash alone, it's the emdash character being used in place of a dash.
The article was obviously generated by ChatGPT.
Everyone thinks they are great at detecting AI slop, but they usually aren't. For art, there are certain giveaways, but for text?
I regularly find myself avoiding the use of the em-dash now even though it is exactly what I should be writing there, for fear of people thinking I used ChatGPT.
I wish it wasn't this way. Alas.
I have a degree in Journalism and now work in customer support. Occasionally, people accuse me of being an AI because of my writing style.
Thankfully, no one I report to internally wants me to simplify my English to prevent LLM accusations. The work I do requires deliberate use of language.
ai slop
False accusations of AI writing are becoming absurd and infuriating.
The other day I saw and argued with this accusation by a HN commenter against a professional writer, based on the most tenuous shred of evidence: https://news.ycombinator.com/item?id=46255049
as a researcher, writing ended up being my job, and more specifically, writing in english. i never developed any sentimental link to the english language, to me it always felt bland, because i had to use it in bland environments, to write texts that had to be bland and manneristic.
chatgpt revolutionized my work because it makes creating those bland texts so much easier and fast. it made my job more interesting because i don't have to care about writing as much as before.
to those who complain about ai slop, i have nothing to say. english was slop before, even before ai, and not because of some conspiracy, but because the gatekeepers of journals and scientific production already wanted to be fed slop.
for sure society will create others, totally idiosyncratic ways to generate distinction and an us vs others. that's natural. but, for now, let's enjoy this interregnum...
Sorry but using the emdash is just a shitty, over corporate way to write, and it instantly rubs some spot in the brain for some people; it doesn't matter if it was generated by an llm or not.
Emdashes can also be part of beautiful writing, like this: https://poets.org/poem/feeling-first
While author is correct in general, I would like to add a counter-point regarding em-dashes specifically. Yes, many people use them like this - and many website frameworks will automatically replace a keyboard not-really-a-minus symbol with em-dash. So that is not a sign of the LLM generated slop.
What LLMs also do though, is use em-dashes like this (imagine that "--" is an em-dash here): "So, when you read my work--when you see our work--what are you really seeing?"
You see? LLMs often use em-dashes without spaces before and after, as a period replacement. Now that is only what an Oxford professor would write probably, I've never seen a human write text like that. So those specific em-dashes is a sure sign of a generated slop.
> What LLMs also do though, is use em-dashes like this (imagine that "--" is an em-dash here): "So, when you read my work--when you see our work--what are you really seeing?"
>You see? LLMs often use em-dashes without spaces before and after, as a period replacement.
It would not make any sense at all to use periods in the places where those em-dashes are supposedly "replacing" periods in the example.
It could also—hear me out here—be me just using compose + --- .
(Not that I used n- or m- dash previously, I used commas, like this! )
But some people learn n- and m-dash, it turns out. Who knew!
Tbh whether I use spaces around em dashes depends more on the font than anything. Some fonts have em dashes that are so long that putting spaces around them would be ridiculous.
[flagged]
Those people broadly fall under "the Oxford professor" catch-all phrase. Obviously. I was talking about 99.99% of random internet texts, which do not conform to any Manual of style and are not written by literature majors. If I see a text authored by some known figure or in a respectable journal/site, then I don't have a task of detecting LLM slop in the first place. But when I do want to know if the text is generated or not, it is usually written by less sophisticated crowd, or anonymous.
The author uses dash (-) not em dash (—), there is a big difference in that everyone has a dedicated dash/undersocore key on their keyboard, but nobody has a em dash key. You can use word processing software etc, but using em dash consistently throughout a text is very unnatural in casual written texts.
There is an easy shortcut for em dashes on macOS, Opt+Shift+-. This makes it really easy to use them, which I do all the time in casual settings (indeed, more often than in formal settings).
And a very easy shortcut in Vim: Ctrl+K M -
Autocomplete does that for me (bilingual English/Portuguese).
I do have an em dash key on the keyboard on my phone :)
Actually, there's a sweet solution to the writing and art crisis we are inflicting ourselves with in our AI craze. I call it "the island". Just find a nice tiny islet somewhere, make a few houses, and rent them by the week to writers/artists. No internet in the place. Rent out sanctioned devices; glorified typewriters without Internet access nor GPU nor CPU fast enough to run an LLM. Bring a notary to certify stuff was purely human-made. Have fun with like-minded individuals.