The epistemic limits of generative AI disclosure in scholarly authorship

Generative AI disclosure is commonly treated as a means of making manuscript production transparent. This article argues that its epistemic value is more limited. Authors may accurately report observable operations, retained outputs, and verification procedures, but functional labels such as “language polishing” or “idea generation” cannot fully reconstruct the causal contribution of AI to a finished manuscript. Nevertheless, such descriptions may strongly influence scholarly evaluation. To examine this asymmetry, the article develops the readerly distraction model, distinguishing justified provenance monitoring from disproportionate attention to features insufficiently related to the disclosed operation. Readerly distraction is analytically distinct from evaluative stigma, although disproportionate monitoring may contribute to negative perceptions of originality, competence, or trustworthiness. The model generates testable predictions concerning attention allocation, comprehension, and evaluation, including comparisons between AI assistance and functionally equivalent human mediation. The article further proposes epistemically proportionate reporting through three complementary categories: methodological reporting, governance reporting, and production-provenance disclosure. Reporting requirements should reflect the role of adopted outputs, their epistemic risks, and the procedures used to verify them rather than the quantity of AI involvement. Finally, the article develops accountable judgment as a standard of authorship comprising understanding, verification, decision authority, and answerability. Unlike nominal responsibility, accountable judgment requires capacities and procedures that can be examined when scholarly claims are challenged. Together, these proposals seek to preserve meaningful transparency without allowing speculative technological provenance to displace the evaluation of evidence and reasoning.

Authors

Publication Details

Journal
Discover Artificial Intelligence
Published
2026-11-19
DOI
https://doi.org/10.1007/s44163-026-02808-6
Primary Topic
Academic integrity and plagiarism
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

The epistemic limits of generative AI disclosure in scholarly authorship

Minoru Matsui
Discover Artificial Intelligence
Academic integrity and plagiarism
article

The epistemic limits of generative AI disclosure in scholarly authorship

Minoru Matsui
article en

Abstract

Generative AI disclosure is commonly treated as a means of making manuscript production transparent. This article argues that its epistemic value is more limited. Authors may accurately report observable operations, retained outputs, and verification procedures, but functional labels such as “language polishing” or “idea generation” cannot fully reconstruct the causal contribution of AI to a finished manuscript. Nevertheless, such descriptions may strongly influence scholarly evaluation. To examine this asymmetry, the article develops the readerly distraction model, distinguishing justified provenance monitoring from disproportionate attention to features insufficiently related to the disclosed operation. Readerly distraction is analytically distinct from evaluative stigma, although disproportionate monitoring may contribute to negative perceptions of originality, competence, or trustworthiness. The model generates testable predictions concerning attention allocation, comprehension, and evaluation, including comparisons between AI assistance and functionally equivalent human mediation. The article further proposes epistemically proportionate reporting through three complementary categories: methodological reporting, governance reporting, and production-provenance disclosure. Reporting requirements should reflect the role of adopted outputs, their epistemic risks, and the procedures used to verify them rather than the quantity of AI involvement. Finally, the article develops accountable judgment as a standard of authorship comprising understanding, verification, decision authority, and answerability. Unlike nominal responsibility, accountable judgment requires capacities and procedures that can be examined when scholarly claims are challenged. Together, these proposals seek to preserve meaningful transparency without allowing speculative technological provenance to displace the evaluation of evidence and reasoning.

Discover Artificial IntelligenceVol. 6(1)
Academic integrity and plagiarism
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon