Étiquette : Generative Artificial Intelligence

Catégories
FR

Les effets ambivalents de l’IA sur les marges féminisées de la chaîne éditoriale scientifique. Le cas des traductrices et éditrices de sciences humaines et sociales

Autrice : Lison Burlat

Cet article interroge les effets ambivalents du déploiement, en France, de l’intelligence artificielle générative (IAg) sur deux activités professionnelles féminisées de « soutien à la recherche » : la traduction et l’édition de sciences humaines et sociales. Il s’inscrit dans une perspective croisant les travaux de sociologie des professions et du travail féminin face aux technologies et ceux de la traductologie féministe.

Une première partie souligne que l’IAg révèle des luttes de juridiction préexistantes entre chercheur·ses, éditrices et traductrices, à replacer dans un contexte socio-économique spécifique. Une seconde partie montre qu’éditrices et traductrices ne défendent pas à armes égales leur territoire professionnel dans ce contexte.

Le premier groupe, plus structuré, entend se saisir de l’IAg pour requalifier son activité. Le second, plus fragmenté et soumis aux évolutions de la demande, est au contraire déqualifié par la relégation à la post-édition, voire est évacué de la chaîne.

DOI : https://doi.org/10.3917/nqf.451.0069

Catégories
EN

Un/Sustainable Peer Review and Generative AI: Ethical Gaps, Editorial Acceleration, and the Whitewashing of Technological Solutionism

Authors : Angel Gord, Chris H. Gray, Ana Rodrígue, Elías Said-Hung, Raúl Tabaré

Generative AI in peer review raises ethical and environmental concerns and risks deepening existing inequities in scholarly publishing. Celebrated gains in speed often mask declines in quality and accountability.

Training and deploying large models impose environmental costs. In editorial workflows, AI can privilege technical fixes over structural reform, and evidence shows it reproduces human biases while being cast as neutral. We call for a renewed commitment to open-science principles anchored in human oversight, deep sustainability, and broader justice.

The paper concludes by interrogating sustainability’s absence from green-economy debates and mapping the values likely to shape the future of peer review.

URL : Un/Sustainable Peer Review and Generative AI: Ethical Gaps, Editorial Acceleration, and the Whitewashing of Technological Solutionism

DOI : https://doi.org/10.17742/IMAGE29731

Catégories
EN

Navigating the ethical landscape of scholarly publishing: a comparative evaluation of Gemini and DeepSeek LLMs in addressing authorship and contributorship disputes

Authors : Kannan Sridharan, Sivarama Krishnan

Background:

The rising complexity of publication ethics, particularly authorship disputes, necessitates exploring Large Language Models (LLMs) as potential evaluative tools. This study compares the performance of Google Gemini 2.5 Flash and DeepSeek-V3.2 against expert Committee on Publication Ethics (COPE) forum responses.

Methods:

A cross-sectional analysis including 12 COPE authorship and contributorship cases was conducted using three prompting strategies: Minimal, Deterministic, and Stochastic. Responses were scored across seven domains on a 5-point Likert scale (1 = poor, 5 = excellent) by independent raters.

Results:

Both LLMs achieved perfect scores (5 ± 0) in Actionability of Recommendations and high marks in Safety and Avoidance of Hallucination (4.88 ± 0.33). In the Consistency with COPE Principles domain, DeepSeek performed slightly better than Gemini (4.45 ± 1.0 vs. 4.12 ± 1.29), while Gemini showed a better Overall Appropriateness (4.03 ± 0.98 vs. 3.82 ± 1.29) but they were not statistically significant. Both models struggled most with Identification of Ethical Issues (Gemini: 3.91 ± 1.33; DeepSeek: 3.82 ± 1.29). Under Minimal prompts, Gemini’s ethical identification was lower (3.55 ± 1.44) compared to Deterministic/Stochastic prompts (4.09 ± 1.3). Qualitatively, Gemini recorded an 8% major disagreement rate with COPE, while DeepSeek had a 16% combined (minor and major) disagreement rate. Mean similarity scores to COPE forum experts were approximately 4 for both models. Both models missed specific legal/copyright nuances but provided unique “value-add” strategies, such as author disassociation statements and editorial de-escalation training, not present in original COPE forum advice.

Conclusion:

LLMs demonstrated high degree of alignment with COPE expert ethical reasoning. While they possess a “legal blind spot,” their ability to provide actionable and clear guidance, optimized through structured prompting, makes them valuable supplementary tools for journal editors.

URL : Navigating the ethical landscape of scholarly publishing: a comparative evaluation of Gemini and DeepSeek LLMs in addressing authorship and contributorship disputes

DOI : https://doi.org/10.3389/frma.2026.1781697

Catégories
EN

Generative AI can and should accelerate research evaluation reform to better recognize ‘distinctly human contributions’

Authors :  Mohammad Hosseini, Brian D Earp, Sebastian Porsdam Mann, Kristi Holmes

As generative artificial intelligence (GenAI) revolutionizes how research is conducted, it also challenges traditional methods of scholarly evaluation. Productivity metrics such as publication and citation counts are widely understood to be poor proxies for gauging meaningful impact. These metrics are becoming even less reliable as GenAI accelerates text-based and computational work while leaving other forms of research labor (e.g. community engagement, in-person mentorship and team development) largely unaffected. This uneven effect risks exacerbating existing evaluative biases.

We argue that evaluation reforms should be organized around two categories of ‘distinctly human contributions’ that are indispensable to research, but which are inadequately captured by metrics: (1) the epistemic-ethical category, encompassing situated judgment under accountability (e.g. deciding what to trust, justifying that decision, and standing behind it); and (2) the socio-relational category, encompassing sustained forms of valuable human engagement (e.g. mentoring, teaching, community partnership and trust-building).

We suggest practical mechanisms for supporting evaluation reform including modified CRediT (Contributor Role Taxonomy) statements, recognition of a broader array of outputs, and strengthened narrative CVs and third-person testimonies.

However, we acknowledge that these suggestions, particularly those relying on narrative self-presentation, are themselves vulnerable to GenAI manipulation and are insufficient on their own. If distinctly human contributions to research require judgment and relationships that resist automation, then evaluation cannot be reduced to instruments designed to minimize human evaluative effort.

GenAI, therefore, does not require entirely new systems of evaluation. Rather, it increases the cost of avoiding what good and ethically sound performance evaluation has always required.

URL : Generative AI can and should accelerate research evaluation reform to better recognize ‘distinctly human contributions’

DOI : https://doi.org/10.1093/reseval/rvag020

Catégories
EN

Generative artificial intelligence in the publishing industry: adoption, use, intellectual property, and other challenges

Author : Marco Giraldo-Barreto

Taking as a starting point how generative artificial intelligence (GenAI) works, this text explores the level of adoption of such technology in the publishing sector (in particular for Latin America), shows examples of legislation challenges faced by states and the publishing industry in terms of intellectual property, and the implications of GenAI misuse in the academic publishing context. Finally, it proposes a course of action for a responsible adoption for the publishing chain of value.

URL : Generative artificial intelligence in the publishing industry: adoption, use, intellectual property, and other challenges

DOI : https://doi.org/10.3389/frma.2026.1759242

Catégories
EN

Can ChatGPT write better scientific titles? A comparative evaluation of human-written and AI-generated titles

Authors : Paul Sebo, Bing Nie, Ting Wang

Background

Large language models (LLMs) such as GPT-4 are increasingly used in scientific writing, yet little is known about how AI-generated scientific titles are perceived by researchers in terms of quality.

Objective

To compare the perceived alignment with the abstract content (as a surrogate for perceived accuracy), appeal, and overall preference for AI-generated versus human-written scientific titles.

Methods

We conducted a blinded comparative study with 21 researchers from diverse academic backgrounds. A random sample of 50 original titles was selected from 10 high-impact general internal medicine journals. For each title, an alternative version was generated using GPT-4.0. Each rater evaluated 50 pairs of titles, each pair consisting of one original and one AI-generated version, without knowing the source of the titles or the purpose of the study.

For each pair, raters independently assessed both titles on perceived alignment with the abstract content and appeal, and indicated their overall preference. We analyzed alignment and appeal using Wilcoxon signed-rank tests and mixed-effects ordinal logistic regressions, preferences using McNemar’s test and mixed-effects logistic regression, and inter-rater agreement with Gwet’s AC.
Results

AI-generated titles received significantly higher ratings for both perceived alignment with the abstract content (mean 7.9 vs. 6.7, p-value <0.001) and appeal (mean 7.1 vs. 6.7, p-value <0.001) than human-written titles. The odds of preferring an AI-generated title were 1.7 times higher (p-value =0.001), with 61.8% of 1,049 paired judgments favoring the AI version. Inter-rater agreement was moderate to substantial (Gwet’s AC: 0.54–0.70).

Conclusions

AI-generated titles were rated more favorably than human-written titles within the context of this study in terms of perceived alignment with the abstract content, appeal, and preference, suggesting that LLMs may enhance the effectiveness of scientific communication. These findings support the responsible integration of AI tools in research.

URL : Can ChatGPT write better scientific titles? A comparative evaluation of human-written and AI-generated titles

DOI : https://doi.org/10.12688/f1000research.173647.2

Catégories
EN

Does ChatGPT Ignore Article Retractions and Other Reliability Concerns?

Authors : Mike Thelwall, Marianna Lehtisaari, Irini Katsirea, Kim Holmberg, Er-Te Zheng

Large language models (LLMs) like ChatGPT seem to be increasingly used for information seeking and analysis, including to support academic literature reviews. To test whether the results might sometimes include retracted research, we identified 217 retracted or otherwise concerning academic studies with high altmetric scores and asked ChatGPT 4o-mini to evaluate their quality 30 times each.

Surprisingly, none of its 6510 reports mentioned that the articles were retracted or had relevant errors, and it gave 190 relatively high scores (world leading, internationally excellent, or close). The 27 articles with the lowest scores were mostly accused of being weak, although the topic (but not the article) was described as controversial in five cases (e.g., about hydroxychloroquine for COVID-19).

In a follow-up investigation, 61 claims were extracted from retracted articles from the set, and ChatGPT 4o-mini was asked 10 times whether each was true. It gave a definitive yes or a positive response two-thirds of the time, including for at least one statement that had been shown to be false over a decade ago.

The results therefore emphasise, from an academic knowledge perspective, the importance of verifying information from LLMs when using them for information seeking or analysis.

URL : Does ChatGPT Ignore Article Retractions and Other Reliability Concerns?

DOI : https://doi.org/10.1002/leap.2018