EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Has AI written this paper? – A case study in „plagiarism detection”

Anna Kamińska

Zagadnienia Informacji Naukowej - Studia Informacyjne · 14.05.2026 · 63 (2A) · 1-16

DOI: 10.36702/zin2025.02.04e ↗

Kurz gefasst

Fachartikel in der Zeitschrift Zagadnienia Informacji Naukowej - Studia Informacyjne, online erschienen am 14.05.2026. Verfasst von Anna Kamińska. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose/Thesis: The paper evaluates the use of GenAI content detectors in the context of verifying academic integrity. The main thesis is that content detectors cannot distinguish AI-generated text from human-authored content that has been stylistically edited only by large language models. Consequently, high detection scores do not constitute reliable evidence of plagiarism. Approach/Methods: A case study method was employed. A biography of a fictional character was created by a human and subsequently edited using ChatGPT and Claude, with precise instructions that limited the AI to a purely editorial role. All versions were analysed using the ZeroGPT detector. Results and conclusion: The original text scored a 0% probability of being AI-generated, while the versions edited by ChatGPT and Claude achieved 96.7% and 100%, respectively. Detectors identify stylistic features characteristic of well-edited texts, rather than the actual origin of the intellectual contribution. Practical implications: The findings indicate the need to revise academic policies in light of AI detectors. Institutions should promote transparent rules for using AI as a supporting tool rather than relying on unreliable technologies that can lead to unfair assessments. Originality/Value: The work provides empirical evidence of the unreliability of AI detectors in the context of plagiarism detection, establishes a clear ethical boundary between using AI as a tool and academic dishonesty, and promotes a conscious approach to new technologies in academia.

Zitieren

APA

Kamińska, A. (2026). Has AI written this paper? – A case study in „plagiarism detection”. Zagadnienia Informacji Naukowej - Studia Informacyjne, 63(2A), 1-16. https://doi.org/10.36702/zin2025.02.04e

Vancouver

Kamińska A. Has AI written this paper? – A case study in „plagiarism detection”. Zagadnienia Informacji Naukowej - Studia Informacyjne. 2026;63(2A):1-16. doi:10.36702/zin2025.02.04e

BibTeX
@article{Kamiska2026,
  author = {Kamińska, Anna},
  title = {Has AI written this paper? – A case study in „plagiarism detection”},
  journal = {Zagadnienia Informacji Naukowej - Studia Informacyjne},
  year = {2026},
  volume = {63},
  number = {2A},
  pages = {1--16},
  doi = {10.36702/zin2025.02.04e}
}
Kostenlos starten →