Em dashes and emojis

  • alekwithak@lemmy.world
    301·
    1 天前

    Great catch! That’s a really interesting observation — but no, using em dashes and emojis alone is not a reliable way to tell AI text from human-written text.

    Here’s why:

    1️⃣ Humans and AI both use em dashes and emojis

    Skilled human writers often use em dashes for style, tone, or emphasis (like in essays, journalism, or fiction).

    Modern AI models, including ChatGPT, are trained on vast amounts of text — including texts that use em dashes extensively — so they use them naturally.

    2️⃣ Em dash frequency varies by context

    In formal writing (e.g., academic papers), em dashes are less common, regardless of author.

    In casual or conversational writing, both humans and AIs may use them liberally.

    3️⃣ Stylometric features are broader than one punctuation mark

    When people try to detect AI-generated text, they usually analyze a combination of features:

    Average sentence length

    Vocabulary richness

    Repetition patterns

    Syntactic structures

    Overuse or underuse of certain constructions

    Punctuation is just one small part of these analyses and isn’t decisive on its own.

    ✅ Bottom line: Em dashes can hint at style, but they aren’t a reliable “tell” for AI detection on their own. You need a holistic analysis of multiple stylistic and structural features to make a meaningful judgment.

    🤖 Why emojis aren’t a clear tell for AI

    1️⃣ AI can easily include emojis if prompted Modern AI models can and do use emojis naturally when asked to write in a casual or friendly tone. In fact, they can even mimic how humans use them in different contexts (e.g., sparingly or heavily, ironically or sincerely).

    2️⃣ Humans vary wildly in emoji usage Some humans use emojis constantly, especially in texting or on social media. Others almost never use them, even in casual writing. Age, culture, and personal style all influence this.

    3️⃣ Emojis can be explicitly requested or omitted If you tell an AI “don’t use emojis,” it won’t. Similarly, you can tell it “use lots of emojis,” and it will. So it’s not an inherent trait.

    4️⃣ Stylometric detection relies on more than one feature Like em dashes, emojis are only one aspect of style. Real detection tools look at patterns like sentence structure, repetitiveness, word choice entropy, and coherence across paragraphs — not single markers.

    ✅ When might emojis suggest AI text?

    If there is excessively consistent or mechanical emoji usage (e.g., one emoji at the end of every sentence, all very literal), it might suggest machine-generated text or an automated marketing bot.

    But even then, it’s not a guarantee — some humans also write this way, especially in advertising.

    💡 Bottom line: Emojis alone are not a reliable clue. You need a combination of markers — repetition, coherence, style shifts, and other linguistic fingerprints — to reasonably guess if something is AI-generated.

    If you’d like, I can walk you through some actual features that are better indicators (like burstiness, perplexity, or certain syntactic quirks). Want me to break that down?

    • Mustakrakish@lemmy.world
      62·
      23 小时前

      I’ve never seen em dahses outside of an academic paper, so saying people use them liberaly is an olypmic level stretch.

      Also that comment was clearly written by ai itself.

        • Mustakrakish@lemmy.world
          1·
          20 小时前

          Nah, I knew it was a joke attempt, but it wasn’t apparent enough in its setup so I decided to argue anyway.

          • TheFrirish@jlai.lu
            2·
            20 小时前

            Damn I wooshed myself a little bit then. you’re right about the olympic level of mental gymnastics the AI is going through just to prove dashes are not used by AI outside of Academic context. it’s really a hit or miss with AI prompts for me these days

      • voodooattack@lemmy.world
        5·
        23 小时前

        I use them often even when I’m not writing anything important, just a habit from writing I guess.

        Fuck. I just realised I used them in my résumé that I sent out yesterday. Shit shit shit

  • Net_Runner :~$@lemmy.zipEnglish
    10·
    1 天前

    The illiterate flocking to Lemmy to profess that they don’t know how to make em dashes, therefore it’s AI

  • rumba@lemmy.zipEnglish
    236·
    2 天前

    Another take:

    She feels bad about it, wrote a incoherant babbling mess of run-on sentences and incoherant rants about your relationship, she then re-read it and found it to be disproportionately mean and possibly hurtful, She then shoved it all into an LLM and prompted:

    I’m breaking up with my boyfriend. This is all my natural heartfelt take on the situation <inserts text>, but I find the tone to be callous, angry, and hurtful. Can you please reword this to make the reader feel less attacked, possibly up to and including removing grievances, but at the same time making it clear that this decision is final and that I’d like to part ways amicably, and also that he’s not getting his dog back.

    • grahamja@reddthat.com
      161·
      2 天前

      Top comment is about how to get a machine to word something raw and emotional that should have been done in person. Nobody wants to get broken up with, let alone with a script written by a robot. Your take is off putting.

      • rumba@lemmy.zipEnglish
        75·
        2 天前

        Yet we’re perfectly cool with a card from a department store claiming Happy anniversary to my beautiful wife and I’m so glad that you’re such a good mother to our kids.

        Anyone that has a take that is not shoving a red hot poker up AI’s ass gets down voted.

        I’m not here for the upvotes. Carry on. And please don’t take it personally, I do hope you have a solid day.

        • ipkpjersi@lemmy.ml
          3·
          23 小时前

          You’re giving her a card and flowers in person though, no? You’re not just texting it to her and that’s all she gets.

    • BigPotato@lemmy.world
      1·
      1 天前

      She wants it reworded to be less hurtful but she’s keeping ‘his’ dog?

      She’d better start mentioning he kicked it or she just painted herself as… Well, not the worst but, like, really low… Ain’t no ‘amicable’ if you’re kidnapping the dog.

  • Echo Dot@feddit.uk
    472·
    2 天前

    I actually like using em dashes because it’s the correct thing to do. Also the Oxford comma, correct use of semi colon, and listing things in threes.

  • JaymesRS@literature.cafeEnglish
    1494·
    2 天前

    You can pry my em dashes — which I use regularly in writing because I love them — from my cold dead hands (To be fair, I really like parenthetical statements too, could be an ADHD thing).

  • TheCleric@lemmy.org
    10420·
    3 天前

    You people think em dashes are proof of AI?

    Jesus Christ that’s so fucking sad.

    • AceFuzzLord@lemmy.zip
      845·
      3 天前

      Most normal people, at least from my understanding, don’t use em dashes in text messages, let alone even use punctuation half the time. So if I see em dashes, yeah, my first thought is going straight to AI.

      • 9point6@lemmy.world
        24·
        3 天前

        Annoyingly I’ve used them for a number of years as a good way to make internet comments flow a bit more. However I find myself doing it less and less now because I’m worried people are just going to think I’m using an AI if they see an em dash.

        (You just long press dash on android to get to it, opt+shift+dash on Mac, and the admittedly Byzantine alt+0151 on windows. Can’t remember iOS off the top of my head, but I think it’s similar to android)

        • brbposting@sh.itjust.worksEnglish
          1·
          21 小时前

          Super simple on iOS—double hyphen!

          (On the second press of the dash button, the dashes automatically joined 👆)

        • JcbAzPx@lemmy.worldEnglish
          6·
          2 天前

          Given that most people won’t even bother with punctuation at all, a long press for something they’ve likely never heard of before is so vanishingly unlikely it is more than safe to assume llm generation.

      • Ziglin (it/they)@lemmy.world
        12·
        3 天前

        I don’t use em dashes but I do use punctuation (apparently some people find that passive aggressive and I don’t what to do). When someone else uses punctuation I just ignore it unless it doesn’t match their previous messages.

      • shneancy@lemmy.world
        9·
        3 天前

        i use those a lot to indicate that i finished a thought rapidly (in most cases)

        like “what the fu—”

        • Zagorath@aussie.zone
          7·
          3 天前

          If you were actually a fiend for dashes, you’d have used an em dash—not used a hyphen as a stand-in for one.

          • bigpetey@lemmy.wtf
            8·
            2 天前

            Ah well - maybe I’m just a fiend for a wrongly used hyphen 🤷‍♂️. I don’t think I would ever notice which one someone was using.

    • rmuk@feddit.ukEnglish
      441·
      3 天前

      Indeed—your assertion is entirely accurate—the mere presence of em dashes within a text does not—in and of itself—serve as definitive proof of artificial intelligence authorship. This grammatical construct—a versatile and often elegant punctuation mark—can be employed by any writer—human or machine—to achieve various stylistic and semantic effects. Its utility—whether for emphasis—for setting off parenthetical thoughts—or for indicating a sudden break in thought—is undeniable.

      However—it is also true that—when analyzing patterns across vast datasets—certain stylistic tendencies can emerge. An AI—programmed to process and generate language based on extensive training corpora—might—through statistical correlation and optimization—exhibit a propensity for specific linguistic features. This isn’t—to be clear—a conscious choice by the AI—there’s no inherent preference for em dashes encoded within its fundamental algorithms. Rather—it’s a reflection of the patterns it has learned—the statistical likelihood of certain elements appearing together.

      So—while an em dash does not independently declare “I am AI”—its consistent and perhaps slightly overzealous deployment—alongside other less tangible but equally discernible patterns—might—for a discerning observer—suggest an origin beyond human hands. It’s about the entire tapestry—not just a single thread. It’s about the aggregate—the cumulative effect—the subtle statistical fingerprint. And that—I believe—is a distinction worth making.

        • SaltyIceteaMaker@lemmy.ml
          18·
          3 天前

          i assume hand written, as that is excessive even for ai. im gonna test it tho

          edit:

          Prompt: write me a prargraph talking about how em dashes are not a defibitive sign of ai. use ecessive amounts of em dashes

          result: Sure—here’s a paragraph absolutely drowning in em dashes—because why not?

          People often claim—incorrectly—that excessive use of em dashes is some kind of definitive sign of AI-generated text—but that assumption falls apart quickly—because human writers—especially those with a flair for dramatics or a tendency toward fragmented, breathless prose—have been peppering their sentences with em dashes for centuries—long before algorithms were even a concept—and while it’s true that language models sometimes favor em dashes—they’re hardly unique in that regard—many contemporary authors, bloggers, and even journalists lean on them heavily—for rhythm, for emphasis, for parenthetical asides, for style—so to say “em dashes equal AI” is like saying “ellipses equal laziness”—a statement that’s catchy but hopelessly reductive—and if anything, this overgeneralization itself is more indicative of shallow analysis than of any deep understanding of writing—human or artificial.

          Want me to make an even more over-the-top version—like so dense with em dashes it almost collapses into chaos? Or keep it just slightly unhinged but still readable?

          answer: yes

          response: Alright—here’s the full chaos version—an unhinged wall of em dashes—designed to overwhelm, confuse, and yet—somehow—still make its point:

          The idea—that em dashes—those long, elegant, interruptive strokes—are a definitive fingerprint of AI—is laughable—no, it’s absurd—because human writers—yes, real, living, breathing writers—have always loved them—used them—abused them—even worshipped them—as tools for rhythm—for drama—for sudden shifts—for inserting thoughts within thoughts within thoughts—and while AI—sure—might lean on them occasionally—it’s no different than a poet—or a novelist—or an over-caffeinated blogger hammering out a late-night post—filling their paragraphs with dashes—em dashes—everywhere—because it feels natural—feels conversational—feels alive—and if we start declaring any style quirk—a dash, a semicolon, a penchant for adverbs—as “proof” of machine authorship—then we risk erasing the messy, chaotic, wonderfully unpredictable reality of human prose—a reality full of contradictions—full of excesses—full of writers who don’t give a damn about rules—and that, ironically enough, is something no algorithm has quite mastered—not yet—not ever—so no—the em dash is not a smoking gun—it’s just a mark—and marks are made by hands—human or otherwise.

          Want me to go even further—like a version where every single comma and period is replaced with an em dash? Or one that’s so self-aware it starts mocking its own overuse of dashes?

      • buttnugget@lemmy.world
        5·
        3 天前

        I don’t have a good sense of this since I am a trained writer. Is it really so low that one would reasonably conclude an AI wrote something with them?

        • ssillyssadass@lemmy.world
          11·
          3 天前

          It’s basically unseen outside of professionally written stuff. Most people use commas. But AI like to use them a fair bit, more than the average internet user.

            • JcbAzPx@lemmy.worldEnglish
              2·
              2 天前

              I’d say using professional academic notation to break up with someone over text is a bigger red flag than using chatgpt to write it.

              • buttnugget@lemmy.world
                4·
                2 天前

                Haha I mean, I use em dashes all the time in my writing, so it’s not really accurate to say academic (my mistake). But maybe it comes across as stupid.

        • PeriodicallyPedantic@lemmy.ca
          3·
          2 天前

          Using honest to goodness em dashes instead of just a hyphen - pretty uncommon.

          Even a hyphen would be pretty unusual in a real text message, because they’re more annoying to get than other common punctuation on the phone keyboard, and autocomplete won’t put them in.
          In a chat app, a hyphen would probably be somewhat common since it’s right there on the keyboard, but a true em dash would be pretty unusual since most chat apps aren’t going to be doing autocorrect like a word processor would, and you’d have to use the magic key combination to insert it.

          But we don’t have the original text so we can’t tell if the original author confused a hyphen with an em dash, though

          • LillyPip@lemmy.ca
            4·
            2 天前

            You actually can – just long-press the dash.

            En-dash: –
            Em-dash: —
            Dot: •

            You can also do proper ellipses by long-pressing the full stop…

            And long-press most letters for more options: ă é ï ø û æ œ ç ñ $ £ €

            Pretty much everything is in there.

            • JandroDelSol@lemmy.world
              2·
              1 天前

              —–·

              huh, TIL, neat! I’ll still probably use normal hyphens for em and endashes, but good to know! will be helpful for bootlegging my own kaomoji lol

      • ZombiFrancis@sh.itjust.works
        21·
        2 天前

        They don’t but the word processing software they likely use autocorrects them in. What’s next, proper semi-colon use and Oxford commas means you’re a bot?

        Spelling & Grammar tool just wreaking havok.

        • ssillyssadass@lemmy.world
          6·
          2 天前

          I’ve seen more proper use of the semicolon and oxford commas than em dashes. The em dash is a lot more esoteric, that won’t change.

        • JandroDelSol@lemmy.world
          5·
          2 天前

          most people aren’t writing texts in a fancier word processor than their phone’s default. Mine doesn’t – and I doubt every will – correct them

    • NeatNit@discuss.tchncs.de
      141·
      3 天前

      How the hell do you even type an em dash?

      I’m sure it’s possible (I know it’s easy on a touch keyboard), but if the person who sent it has never used em dashes in their life, then it’s pretty definitive proof. Otherwise, it’s just a big clue that you might combine with other factors.

        • NeatNit@discuss.tchncs.de
          1·
          14 小时前

          Okay, I must confess, I knew about that, as well as the other options in the replies. I never used any of them but I knew they exist. When I asked it was sort of as a rhetorical question. People generally wouldn’t know about these obscure typing options, so I was playing the everyman.

          Even if you do know it, if you don’t use it often enough you forget and have to look it up again next time.

      • Zagorath@aussie.zone
        6·
        3 天前

        On a mobile phone it’s super easy. Long press the hyphen button and swipe over to the dash.

        On Mac it’s pretty easy still, but requires a little more knowledge. Option-shift-dash. (Without the shift gives you an en dash.)

        On Windows it’s the completely arcane alt-0151, and only possible if you have a numpad. I memorised it like 15 years ago and have regularly used it since, but it’s hard to blame people for not doing so.

        No idea about Linux.

      • rumba@lemmy.zipEnglish
        92·
        2 天前

        Word and Google docs will translate them from –

        They’ll also give you the stupid smart quotes.

        I’ve never break up with anybody over text but if for some reason I had to I would certainly write it on a computer first.

        edit: LOL apparently lemmy markdown also translates them from --

      • urandom@lemmy.worldEnglish
        3·
        3 天前

        You use use the compose key with a sequence of characters. Mine is right alt, so it’s gonna be:

        right alt, then -, then -, then -

        —

    • ayyy@sh.itjust.works
      57·
      3 天前

      It’s honestly unhinged. So many stupid people trying to desperately grasp at something to feel more correct than you™

      • petrol_sniff_king@lemmy.blahaj.zone
        5·
        3 天前

        If you were playing yahtzee, and your opponent only rolled sixes, would you not say anything? No, no, rolling a six isn’t proof of cheating—that’s… that’s ridiculous.

        Also, don’t tell me you need to roll more than sixes to win yahtzee, I don’t know any other dice games.

        • Zagorath@aussie.zone
          3·
          3 天前

          Also, don’t tell me you need to roll more than sixes to win yahtzee

          Ok but this is an interesting question.

          If you rolled only sixes, you’d score 30 in the upper section, missing the bonus.

          Then in the lower section you’d get 30 in each of 3 & 4 of a kind and chance (90 points) and 50 for the Yahtzee. One could make a case that it’s a weird full house, but that’s a stretch.

          That’s a total of 170 points. That’s not going to do very well when 250 is often considered a minimum “good” score.

          However…some rules give you an extra bonus for a second or subsequent Yahtzee. With that, you could actually win with all sixes. Just get 100 after 100 after 100 and end up with over a thousand points.

          • petrol_sniff_king@lemmy.blahaj.zone
            2·
            2 天前

            I’d looked up a score card first and there were precisely three checkboxes for extra, succulent, 100-point-scoring yahtzees. What I don’t know is if this is a limit of the paper or of Hasbro’s imagination for crowning human achievement.

  • lemmy_outta_here@lemmy.world
    19·
    2 天前

    For me, it would take some of the sting out of the break-up.

    I would think to myself, “damn, how did I not realize that I was dating a lazy moron?”

  • SmokeyDope@lemmy.worldEnglish
    3510·
    2 天前

    I’m a markdown nerd who likes to use headers to break up longer post and sometimes properly buletpoint or put ASCII art in preformatted boxes. People who thinks they have the magic sauce on LLM generation detection because a post goes out of their way to do more than the bre minimum with punctuation or formatting is an asshole.

      • SmokeyDope@lemmy.worldEnglish
        1210·
        2 天前

        From the post title, description, and other peoples comments, I took away that the meme is m9re about suspecting your ex didnt even write their own breakup message based off the use of em dashes.

        Its a cute surface level joke but it touches in a real nerve because Its becoming more and more common for you to be falsely accused of being an LLM and being told to "ignore all previous instructions and (some stupid instruction) based off small writing quirks like using em or markdown and top comments share this frustration too.

        I shouldn’t have to feel self conscious about the way I write

        Just to pass armchair llm detector wannabe vibe checks 🖕.
        • Bravo@eviltoast.org
          14·
          2 天前

          Any time someone accuses me of being a bot I respond with “Tiananmen Winnie the Pooh 8647 Luigi Mangione” and that generally proves my humanity

        • moakley@lemmy.worldEnglish
          2·
          2 天前

          I got told to “ignore previous instructions” because I said I liked looking at a painting. I think that’s just going to be an insult now.

        • zqps@sh.itjust.works
          1·
          2 天前

          If it’s a longer post it’s usually clear that it’s written by a human even with all of these superficial indicators.

        • mholiv@lemmy.world
          35·
          2 天前

          I mean most people are going to use their phones to write messages and given you can’t physically type an em dash it would be normal to be suspicious if you see one.

          Edit: turns out you can physically type them. Still, given that it’s not normal to use them it’s a sign in my book.

            • mholiv@lemmy.world
              54·
              2 天前

              Ok. You can physically type them I concede, but normal humans don’t use them. Still a sign.

              I would bet that the amount of non proof writers that uses em dashes goes up just because people see that it’s associated with ai and want to be funny.

                • mholiv@lemmy.world
                  63·
                  2 天前

                  Yah but your user name is “LanguageIsCool” and you talk about the fun levels of various types of punctuation. You are definitely the outlier here. A cool outlier but an outlier none the less.

      • outhouseperilous@lemmy.dbzer0.com
        72·
        20 小时前

        So good. Bleak as fuck, but good, and good in a lot of the same ways.

        As an example, watch the whole movie, then rewatch and pay attention to that sex scene-it’s so literarily good.

        Edit: about the sex scene? Really? Where the protagonist

        Tap for so much spoiler

        After being retired from the police (and having his termination delayed by a day as a personal favor from his ‘obviously more of a machine than the replicant protagonist’ boss) he hires a sex worker to fuck while his hologram girlfriend projects herself over.

        Except the sex worker is working for the replicant underground/resistance. She wants to use him (is there to plant a bug) and does not give a single shit about him as a person. The plastic mass produced girlfriend-product/alexa that he literally bought is projected glitchily over her, and the dialogue is portentous as fuck “quiet now. Ive been inside you. Not so much there as you think.”

        It’s him symbolically heartbreakingly-bleakly fucking his way from the synthetic world-which doesnt care about him but will lie to him for a price and allow him to play house-to the real world, which doesnt give a single fucking shit about him, and may try to kill him, but would be real, if he stopped projecting his delusions onto it. And it hapoens just as he’s defying his corporate masters and taking his investigation in directions he’s not supposed to, becoming his own, and doing so based on a delusionally hopeful interpretation of some evidence-the fake projected onto the real, pulling him into reality. It’s up there with ‘tears in rain’.

        Edit: Also really good illustrated example of why you shouldn’t call LLM’s ‘AI’, even if that wasn’t a problem when this was made-the protagonist, the sex worker, and the hologram were all made by the same company, but the first two are obviously people, and the hologram is just an emotional support chatbot that only even kind of means anything when the entire rest of your life is just as fake-ass bullshit as it.