AI evidence extraction

Plagiarism Ethics in Higher Education and Scholarly Publishing: A Systematic Literature Review of Behaviors, Policies and AI-Driven Challenge

Maskur Maskur, Didik Dwi Prasetya, Hakkun Elmunsyah, Siti Sendari · International Journal of Pedagogical Humanities and Social Studies · 2026

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

9/10

Relevance

1/4

Quality (LMQS)

Evidence

Citations

0.00

FWCI

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.66324/ijphss.v1i3.162

Methodology & findings

Study design

Systematic Literature Review (SLR) adhering to PRISMA criteria.

Sample

N = 26, 3 groups

Primary method

No quantitative statistical methods were employed. This is a qualitative systematic review using narrative synthesis. The review used PRISMA guidelines for transparent reporting of systematic review processes (screening, selection, data extraction stages documented in PRISMA flowchart). Thematic analysis and qualitative theme synthesis were conducted to identify, categorize, and interpret key results organized into three primary categories: Behaviors, Policies, and AI-Driven Challenges.

Main result

The study found that "academic integrity is still a key tenet in the face of changing obstacles" and that "Students' plagiarism, which frequently takes the form of copying and pasting, is essentially encouraged by the accessibility of digital access." Additionally, the literature demonstrates that "plagiarism is often common due to a lack of awareness about how to avoid it, rather than purely due to deliberate intent." The authors note that generative AI presents unique challenges, as "the issue underwent a significant transformation" with the emergence of ChatGPT between 2023-2025, introducing "the 'digital erosion of intellectual integrity,' which is thought to be more complicated than traditional copying due to the blurring of the boundaries between fraud and technological innovation."

Reports effect sizes.

Research paradigm

qualitative synthesis; mixed-methods interpretive

Author conclusions

The authors conclude that "academic ethics and integrity are dynamic issues that go beyond the conventional notion of plagiarism" and that "plagiarism seems as a complicated problem requiring publication retractions, yet it is primarily motivated by ignorance and academic pressure rather than totally malevolent intent." They emphasize that "the biggest threat comes from generative AI tools, which have caused a 'digital erosion of intellectual integrity' by casting doubt on ideas of authorship, copyright, and the efficacy of conventional plagiarism detection software." The authors advocate for institutional change: "higher education institutions must adopt an integrative decision-making model, provide a clear roadmap for the responsible use of AI, and move from a primary focus on detection to a pedagogy of AI ethics."

Risk of bias

Language restriction (English and Indonesian only) - may exclude relevant non-English literature; Temporal restriction (2020-2025) - potential for recency bias; Title-based search constraint - may exclude relevant articles addressing concepts in abstract/keywords only; Incomplete retrieval of reports (17 of 58 potentially eligible reports not obtained) - missing data bias; Single-stage screening for initial 214 records with automatic flagging (68 records) and unclear removals (52 records) - potential selection bias; No inter-rater reliability reported for screening and eligibility decisions; Narrative synthesis approach without quantitative pooling - subjective interpretation risk; Language bias: restriction to English and Indonesian publications only; Publication bias: reliance on published literature in indexed databases; Selection bias: title-field restriction initially excluded relevant abstract/keyword content; Reporting bias: focus on 26 included papers from 214 initial records (88% attrition); Methodological heterogeneity: mix of study designs (case studies, research articles, reviews) without formal quality assessment tool application documented; Selection bias: Title-only search constraint may have excluded relevant literature in abstracts and keywords; Language bias: English and Indonesian language restriction only; Publication bias: Articles not fully retrieved (n=17) during full-text screening phase; Temporal bias: 2020-2025 publication window limits historical perspective; Screening bias: Records flagged as ineligible by automatic program (n=68) and 52 records eliminated for unclear reasons prior to formal screening

Limitations

The authors note that "Although the title search produced very accurate results, the researchers realized that this limitation would leave out significant literature in the abstract or keyword fields that addressed behaviors, policies, and AI-driven concerns." Additionally, the review faced retrieval limitations: "after 17 reports were not fully retrieved (Reports not obtained)," only 41 of 58 potentially eligible reports could be fully evaluated
The temporal restriction (2020-2025) and language constraints (English and Indonesian only) may have excluded relevant publications
The review's reliance on qualitative theme synthesis rather than meta-analysis limits quantitative synthesis of effect sizes.

Open questions raised

The review identifies gaps including: (1) limited understanding of how policies adapt to emerging technologies; (2) insufficient research on psychological and mental health implications on educators; (3) need for comprehensive examination of AI function in editorial administration of scientific journals; (4) lack of integrative frameworks for responsible AI adoption in higher education; (5) shortage of pedagogical approaches to teaching AI ethics versus detection-focused policing; (6) research gap regarding global policy harmonization and international ethical norm disparities.
Limited research on institutional capacity to balance preventing technology exploitation with capitalizing on AI's innovative prospects
Gap in policies addressing generative AI ethical issues (noted as 'technology gap')
Insufficient framework for responsible AI adoption in higher education institutions
Need for integrative decision-making models combining AI with ethics education
Limited research on psychological and mental health implications on educators regarding AI integration

Extracted from: pdfAgreement 64%

Explore related topics

Citation Integrity Research Integrity Meta-Analysis