StoryScope: Investigating idiosyncrasies in AI fiction
arxiv.org
As AI-generated fiction becomes increasingly prevalent, questions of authorship and originality are becoming central to how written work is evaluated. While most existing work in this space focuses on identifying surface-level signatures of AI writing, we ask instead whether AI-generated stories can be distinguished from human ones without relying on stylistic signals, focusing on discourse-level narrative choices such as character agency and chronological discontinuity. We propose StoryScope, a pipeline that automatically induces a fine-grained, interpretable feature space of discourse-level narrative features across 10 dimensions. We apply StoryScope to a parallel corpus of 10,272 writing prompts, each written by a human author and five LLMs, yielding 61,608 stories, each ~5,000 words, and 304 extracted features per story. Narrative features alone achieve 93.2% macro-F1 for human vs. AI detection and 68.4% macro-F1 for six-way authorship attribution, retaining over 97% of the performance of models that include stylistic cues. A compact set of 30 core narrative features captures much of this signal: AI stories over-explain themes and favor tidy, single-track plots while human stories frame protagonist' choices as more morally ambiguous and have increased temporal complexity. Per-model fingerprint features enable six-way attribution: for example, Claude produces notably flat event escalation, GPT over-indexes on dream sequences, and Gemini defaults to external character description. We find that AI-generated stories cluster in a shared region of narrative space, while human-authored stories exhibit greater diversity. More broadly, these results suggest that differences in underlying narrative construction, not just writing style, can be used to separate human-written original works from AI-generated fiction.

Abstract page for arXiv paper 2604.03136: StoryScope: Investigating idiosyncrasies in AI fiction

  • doben@lemmy.mlEnglish
    11·
    2 days ago

    Thankfully, I am not your pupil. If you’re not able to make an argument without sending people off to read into your own philosophies, you ultimately failed to make an argument.

    You’re’ making a categorical error, as this thread is about an Automaton having direct, tangible real-world consequences for people’s live trajectory. It’s not about the legitimacy of the educational system itself, not about the relevance of grades. If it was, I would probably even agree with you.

    Now, that your initial, ideological take has been rebuked after the very first try, you keep arguing a strawman from a privileged position of survivorship bias, with an outdated anecdote and an absolute refusal to acknowledge differential vulnerability, by reducing your pompous deliberations about systemic rot into individual failure. Effectively a cruel form of victim-blaming disguised as enlightenment. And probably one of the most US-American things to do.

    But what is it now, the failure of the system or of the individual? Because your argument seems to be condemning the system as irredeemably corrupt, yet simultaneously you blame individuals for not transcending it. If the system is so rotten, how can individual failure to escape it be a moral failing? Of course you can’t escape that contradiction without exposing your very own misanthropic disgust for those that didn’t manage.

    You, the irrelevant comet in my orbit, only drift further off into the void now, by distorting the problem of a 7% false-positive rate on a life-altering accusation into a one in a million (0.0001%) chance of terrible application of technology. Mixed with unfounded empirical claims about the material conditions about the labor market in the 21 century and spiteful jabs at potential live choices of the ones falling through the cracks.

    Your attempts to claim authority are merely furthering your illegitimacy as a serious person. You’re dismissed. Off you go, lil’ irrelevant comet.

    • MangoCats@feddit.itEnglish
      1·
      1 day ago

      It’s not about the legitimacy of the educational system itself

      Isn’t it? If a bunch of loan sharks are suddenly impeding their borrowers ability to pay because their borrowers have used AI, is AI the problem? As a society, the British empire and colonies transcended debtors prisons some centuries ago, as counterproductive as public hangings of pickpockets.

      rebuked after the very first try

      In your head.

      victim-blaming

      The victims aren’t being victimized by the University’s use of (insert new thing here), they’ve been victimized for decades by overpriced education coupled with a limp job market.

      system as irredeemably corrupt

      Is it not?

      the system is so rotten, how can individual failure to escape it be a moral failing?

      The individuals are flocking to the rotten system which gives little benfit to most, negative benefits to many, and now it’s threatening to throw them out for embracing a future that the rotten system can’t keep up with. This isn’t victim blaming, it’s redirecting gasoline soaked refugees away from a burning building advertising itself as a safe shelter.

      a 7% false-positive rate on a life-altering accusation

      Show me the university that’s taking these false life-altering actions - where’s the list of expelled students? That’s going to turn into a list of fat lawsuits and payouts if it exists, which it doesn’t, because the Universities would never expel their golden geese. This is hot-air threats just as every “academic integrity” bluster of the past 50 years has been.

      Your attempts to claim authority

      I’m no authority, I’m a single column of text on the internet, yet you react as if you are threatened…