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

  • MangoCats@feddit.itEnglish
    1·
    4 hours ago

    I had student loans, luckily they only amounted to 4-5 months’ gross income - lucky because scholarships stepped up and I only had to take a loan for one year instead of all six.

    none of the credentials required for jobs

    I got news for you: you can have all the credentials in the world available from the best Universities, and when a place isn’t hiring, they aren’t hiring. I’ve gone job hunting in Melbourne Florida after Columbia burnt up on re-entry, the entire town wasn’t hiring - that’s not about me, they weren’t hiring anyone.

    I interviewed in Manhattan with a fresh Honors degree from a good school, the employment agents there dumped all over me about how worthless I was, until… I wandered into an office where the interviewer had a glass table, she looked down through it and saw my shoes were freshly shined, my silk tie was bright red ($2 from a street vendor), and I had a goofy-trendy $50 haircut that was in-style, she gushed: “oh, I know exactly who will hire you: Kidder Peabody, you’ll so fit right in.” She barely glanced at my resume’ “you do computer stuff, right? Yeah, they’re always looking for more computer guys, they’re just gonna love you!” Transcripts were never mentioned.

    After .com hit, any remotely computer related degree, straight out of school was a ticket to more than double what I made starting out with the best computer degree for .com, those kids left school making more than my boss with 25 years experience was making 5 years’ earlier.

    Market gets saturated, hiring stops again. It’s different in different fields, but still much the same: cycles. Personal referrals beat the HR screening circus. The place I work today, I had applied to 4 times in the previous 15 years, HR never even called me back. Personal referral got me a work from home spot at a company this company ended up buying, since they bought me I bypassed the HR filters. Once I was past the HR filters they offered my an 8 month salary signing bonus to stay with them at least 1 year, my base pay has increased faster than inflation for 12 years straight now, and the annual bonus runs about 2 months salary.

    I’m not revealing any useful “secrets” here… truth is: the system sucks, when it’s not blowing chunks. But, that’s the world - and trust me when I tell you: academics A) don’t know everything about how the rest of the world works, and B) what they do know they tend to distort for their students in various ways, for various reasons.

    Any rant against academia this long deserves this link: https://rmst202.sites.olt.ubc.ca/files/2022/04/illich_deschooling-society.pdf