Source Details

{
    "source_type": "extension",
    "content_hash": "267b9d1cfec06c930629d2706a27cfacb84fb28c2e8dbda975f98cd21fe848b2",
    "submitted_via": "chrome_extension",
    "extension_version": "1.0.18",
    "original_url": "https:\/\/arxiv.org\/abs\/2508.00238?emc=edit_hh_20260925&nl=well&segment_id=227037",
    "parsed_content": "Computer Science > Computation and Language\n \n \n arXiv:2508.00238 (cs)\n \n \n \n [Submitted on 1 Aug 2025]\n Title:Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English\n Authors:Bryce Anderson, Riley Galpin, Tom S. Juzek View a PDF of the paper titled Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English, by Bryce Anderson and 2 other authors\n View PDF\n HTML (experimental)\n \n Abstract:In recent years, written language, particularly in science and education, has undergone remarkable shifts in word usage. These changes are widely attributed to the growing influence of Large Language Models (LLMs), which frequently rely on a distinct lexical style. Divergences between model output and target audience norms can be viewed as a form of misalignment. While these shifts are often linked to using Artificial Intelligence (AI) directly as a tool to generate text, it remains unclear whether the changes reflect broader changes in the human language system itself. To explore this question, we constructed a dataset of 22.1 million words from unscripted spoken language drawn from conversational science and technology podcasts. We analyzed lexical trends before and after ChatGPT's release in 2022, focusing on commonly LLM-associated words. Our results show a moderate yet significant increase in the usage of these words post-2022, suggesting a convergence between human word choices and LLM-associated patterns. In contrast, baseline synonym words exhibit no significant directional shift. Given the short time frame and the number of words affected, this may indicate the onset of a remarkable shift in language use. Whether this represents natural language change or a novel shift driven by AI exposure remains an open question. Similarly, although the shifts may stem from broader adoption patterns, it may also be that upstream training misalignments ultimately contribute to changes in human language use. These findings parallel ethical concerns that misaligned models may shape social and moral beliefs.\n \n \n \n \n Comments:\n Accepted at AIES 2025. To appear in the AIES Proceedings. 14 pages, 2 figures, 2 tables. Licensed under CC BY-SA 4.0\n \n Subjects:\n \n Computation and Language (cs.CL); Artificial Intelligence (cs.AI)\n \n MSC classes:\n 68T50\n \n \n ACM\u00a0classes:\n I.2; I.2.7\n \n Cite as:\n arXiv:2508.00238 [cs.CL]\n \n \n \n (or \n arXiv:2508.00238v1 [cs.CL] for this version)\n \n \n \n \n https:\/\/doi.org\/10.48550\/arXiv.2508.00238\n \n \n Focus to learn more\n \n \n \n arXiv-issued DOI via DataCite\n \n \n \n \n \n \n Submission history From: Bryce Anderson [view email] [v1]\n Fri, 1 Aug 2025 00:47:33 UTC (275 KB)\n \n \n \n \n Full-text links:\n Access Paper:\n \n \nView a PDF of the paper titled Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English, by Bryce Anderson and 2 other authorsView PDFHTML (experimental)TeX Source\n \n \n \n view license\n \n \n \n Current browse context:\n cs.CL\n \n <\u00a0prev\n \n | \n next\u00a0>\n \n \n new\n | \n recent\n | 2025-08\n \n Change to browse by:\n \n cs\n cs.AI\n \n \n \n \n References & Citations\n \n NASA ADSGoogle Scholar\n Semantic Scholar\n \n \n \n export BibTeX citation\n Loading...\n \n \n BibTeX formatted citation\n \u00d7\n \n \n loading...\n \n \n Data provided by: \n \n \n \n Bookmark\n \n \n \n \n \n \n \n \n Bibliographic Tools\n \n Bibliographic and Citation Tools\n \n \n \n \n \n \n Bibliographic Explorer Toggle\n \n \n \n Bibliographic Explorer (What is the Explorer?)\n \n \n \n \n \n \n \n Connected Papers Toggle\n \n \n \n Connected Papers (What is Connected Papers?)\n \n \n \n \n \n \n Litmaps Toggle\n \n \n \n Litmaps (What is Litmaps?)\n \n \n \n \n \n \n \n scite.ai Toggle\n \n \n \n scite Smart Citations (What are Smart Citations?)\n \n \n \n \n \n \n \n \n \n Code, Data, Media\n \n Code, Data and Media Associated with this Article\n \n \n \n \n \n \n alphaXiv Toggle\n \n \n \n alphaXiv (What is alphaXiv?)\n \n \n \n \n \n \n \n Links to Code Toggle\n \n \n \n CatalyzeX Code Finder for Papers (What is CatalyzeX?)\n \n \n \n \n \n \n \n DagsHub Toggle\n \n \n \n DagsHub (What is DagsHub?)\n \n \n \n \n \n \n \n \n GotitPub Toggle\n \n \n \n Gotit.pub (What is GotitPub?)\n \n \n \n \n \n \n \n Huggingface Toggle\n \n \n \n Hugging Face (What is Huggingface?)\n \n \n \n \n \n \n \n ScienceCast Toggle\n \n \n \n ScienceCast (What is ScienceCast?)\n \n \n \n \n \n \n \n \n \n \n \n Demos\n \n Demos\n \n \n \n \n \n \n Replicate Toggle\n \n \n \n Replicate (What is Replicate?)\n \n \n \n \n \n \n \n Spaces Toggle\n \n \n \n Hugging Face Spaces (What is Spaces?)\n \n \n \n \n \n \n \n Spaces Toggle\n \n \n \n TXYZ.AI (What is TXYZ.AI?)\n \n \n \n \n \n \n \n \n Related Papers\n \n Recommenders and Search Tools\n \n \n \n \n \n \n Link to Influence Flower\n \n \n \n Influence Flower (What are Influence Flowers?)\n \n \n \n \n \n \n \n Core recommender toggle\n \n \n \n CORE Recommender (What is CORE?)\n \n \n \n \n \n Author\n Venue\n Institution\n Topic\n \n \n \n \n \n \n \n \n \n \n \n \n \n About arXivLabs\n \n \n \n \n arXivLabs: experimental projects with community collaborators\n arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.\n \n \n \n \n \n \n \n \n Which authors of this paper are endorsers? |\n Disable MathJax (What is MathJax?)",
    "author_name": "Bryce Anderson, Riley Galpin, Tom S. Juzek",
    "site_name": "arXiv",
    "section": "Computation and Language",
    "publisher": null,
    "ai_headline": "Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English",
    "ai_simplified_title": "Researchers find AI-associated language increasing in unscripted spoken English",
    "ai_excerpt": "This paper investigates the influence of Large Language Models (LLMs) on human language, specifically in unscripted spoken English. Analyzing 22.1 million words from podcasts, the study found a significant increase in LLM-associated words post-2022, suggesting a convergence between human and AI language patterns.",
    "ai_subject_tags": [
        "Large Language Models",
        "Language Change",
        "AI Influence",
        "Computational Linguistics",
        "Spoken English",
        "Lexical Analysis"
    ],
    "ai_context_type": "Research",
    "ai_context_details": {
        "tone": "analytical",
        "perspective": "academic",
        "audience": "specialized",
        "credibility_indicators": [
            "peer_reviewed",
            "data_cited",
            "academic_authors"
        ]
    },
    "ai_source_vector": null,
    "ai_confidence_score": 0.9999999999999999,
    "ai_extraction_metadata": {
        "extracted_at": "2026-09-25T01:09:42.377451Z",
        "ai_model": "gemini-2.5-flash",
        "extraction_method": "automated",
        "content_length": 5459,
        "metadata_prompt_body_chars": 5459,
        "metadata_prompt_truncated": false,
        "url": "https:\/\/arxiv.org\/abs\/2508.00238",
        "existing_metadata": {
            "author_name": null,
            "published_at": null,
            "domain_name": "arxiv.org",
            "site_name": null,
            "section": null,
            "publisher": null
        },
        "openai_refusal_gemini_fallback": false
    }
}

{
    "extracted_at": "2026-09-25T01:09:42.377451Z",
    "ai_model": "gemini-2.5-flash",
    "extraction_method": "automated",
    "content_length": 5459,
    "metadata_prompt_body_chars": 5459,
    "metadata_prompt_truncated": false,
    "url": "https:\/\/arxiv.org\/abs\/2508.00238",
    "existing_metadata": {
        "author_name": null,
        "published_at": null,
        "domain_name": "arxiv.org",
        "site_name": null,
        "section": null,
        "publisher": null
    },
    "openai_refusal_gemini_fallback": false
}