AI evidence extraction

AI In Academic Publishing for Non‐Native English Speakers: The Good, the Bot, and the Ugly

Talip Gönülal, Ramazan Güçlü, Salih Güçlü · Learned Publishing · 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/10.1002/leap.2070

Methodology & findings

Study design

Mixed-methods convergent parallel design examining how non-native English-speaking academics utilize AI tools, their perceived benefits, and concerns regarding AI's influence on academic publishing.

Sample

N = 105

Main result

The study identified three key dimensions of the perceived impact of AI: "the good, reducing linguistic inequalities by improving paper quality and decreasing language-related challenges; the bad, involving inaccurate or misleading AI suggestions, over-reliance on AI tools, and diminished engagement with manuscripts; and the ugly, characterized by failure to disclose AI use, lack of clear guidelines for responsible AI integration in research, homogenization of academic writing, and the emergence of new forms of inequality."

Reports effect sizes.

Research paradigm

Mixed methods (convergent parallel design combining qualitative and quantitative approaches)

Author conclusions

The study "concluded with several recommendations for individual researchers, academic institutions, and publishers and journals to promote the ethical and effective use of AI in academic publishing."

Risk of bias

Selection bias: participants self-selected to participate (recruitment mechanism not detailed); Self-report bias: reliant on participant perceptions of AI benefits and concerns; Language background diversity may introduce heterogeneity in AI tool accessibility and effectiveness; Selection bias: Participants self-selected or were convenience sampled (not specified in abstract); Potential confirmation bias: Researchers may have sought responses confirming AI impact categories; Self-report bias: Reliance on participant perceptions rather than objective measures; Lack of control group: No comparison with non-AI-using researchers; Language representation bias: 25 language backgrounds across 105 participants suggests unequal representation; Selection bias: self-selection of participants willing to discuss AI use; Social desirability bias: participants may underreport or overreport AI reliance; Language background diversity without stratified analysis specification

Open questions raised

The study implicitly identifies gaps in guidelines for responsible AI integration in research and in institutional frameworks governing AI disclosure in academic publishing.
The abstract implies need for clear guidelines for responsible AI integration in research and clearer disclosure requirements for AI use in academic publishing.
Need for clear guidelines for responsible AI integration in research; need for mechanisms to prevent homogenization of academic writing; need for oversight to prevent emergence of new forms of inequality in academic publishing

Data: not_statedCode: not_statedExtracted from: pdfAgreement 64%

Explore related topics

AI Manuscript Writing Academic Writing Mixed Methods