{
  "abstract": "Version 2.3 of Adversarial Multi-AI Falsification is a bounded case study of AI-assisted scholarly workflow governance. It distinguishes selected public verbatim excerpts and message/line maps from author-reported recovery findings and hashes identifying internal artifacts. The public deposit does not contain the full captured transcripts or executed rubric.\n\nFour process-failure types required exclusion or downgrading: semantic drift with unsupported process certification; assignment-scope expansion; prior-task substitution; and unverified or misdescribed legal citations. E2 remains partial. The principal execution/adjudication sequence remains author-reconstructed, and panel unanimity cannot be independently recomputed from the public deposit. The operational proposition remains untested; this revision does not establish it as true.\n\nThe paper proposes a three-field outcome record and six workflow controls, including a two-turn assignment-echo checkpoint. These are unvalidated design propositions, not demonstrated interventions. This is a case report and protocol demonstration, not a benchmark, product comparison, prevalence estimate or complete reproducibility study.\n\nAMAF_v2_3.pdf is the current controlling manuscript, with AMAF_v2_3.docx as its editable counterpart. All seven v2.2 files are retained unchanged as historical records, including the old preparation-candidate manuscript, README, manifest and checksum list. Read 00_README_AMAF_v2_3_RELEASE.txt for the current controlling-version guide; AMAF_v2_3_SHA256SUMS.txt lists this release's file identities. The historical v1.8 paper is retained only for its Section C dual-track summary. No additional private records, new experiment or new empirical validation are released.\n\nGenerative AI systems were used for structured literature discovery, comparative issue mapping, adversarial review, source-check planning, language editing and revision support. They are not authors. The human author retains responsibility for the sources, interpretations, arguments and conclusions.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": null,
  "author_total": 1,
  "authors": [
    {
      "name": "QianJun Yu",
      "orcid": "0009-0007-7071-8965"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/brc8c85mh3d7g.bib",
  "cited_by": 0,
  "doi": "10.5281/zenodo.23489600",
  "fulltext_url": null,
  "id": "brc8c85mh3d7g",
  "issns": [],
  "landing_url": "https://zenodo.org/records/23489600",
  "language": "eng",
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [],
  "openalex": null,
  "pdf_url": "https://zenodo.org/api/records/23489600/files/Adversarial_Multi_AI_Falsification_v1_8_CLEAN.pdf/content",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-19",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/brc8c85mh3d7g.ris",
  "sources": [
    {
      "first_seen": "2026-11-02T12:20:43Z",
      "native_id": "23489600",
      "source": "zenodo",
      "url": "https://zenodo.org/records/23489600"
    }
  ],
  "subjects": [],
  "title": "Adversarial Multi-AI Falsification in Research Workflows: A Bounded Case Study of Scope Error, Verification Failure, and Human Adjudication",
  "title_text": "Adversarial Multi-AI Falsification in Research Workflows: A Bounded Case Study of Scope Error, Verification Failure, and Human Adjudication",
  "type": "preprint",
  "updated_at": "2026-11-02T12:20:43Z",
  "url": "https://research.draconitic.com/paper/brc8c85mh3d7g",
  "venue": "Zenodo",
  "versions": []
}