Do teachers spot AI? Evaluating the detectability of AI-generated texts among student essays
Open accessAbstract
The potential application of generative artificial intelligence (AI) in schools and universities poses great challenges, especially for the assessment of students’ texts. Previous research has shown that people generally have difficulty distinguishing AI-generated from human-written texts; however, the ability of teachers to identify an AI-generated text among student essays has not yet been investigated. Here we show in two experimental studies that novice (N = 89) and experienced teachers (N = 200) could not identify texts generated by ChatGPT among student-written texts. However, there are some indications that more experienced teachers made more differentiated and more accurate judgments. Furthermore, both groups were overconfident in their judgments. Effects of real and assumed source on quality assessment were heterogeneous. Our findings demonstrate that with relatively little prompting, current AI can generate texts that are not detectable for teachers, which poses a challenge to schools and universities in grading student essays. Our study provides empirical evidence for the current debate regarding exam strategies in schools and universities in light of the latest technological developments.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Human users often fail to recognise AI-generated text and are generally overconfident in their ability to do so.
“Previous studies suggested that human users often fail to recognise AI-generated text and are generally overconfident in their ability to do so (Casal & Kessler, 2023;Fleckenstein et al., 2024).”
From “ <scp>ChatGPT</scp> can make mistakes. Check important info.” Epistemic beliefs and metacognitive accuracy in students' integration of <scp>ChatGPT</scp> content into academic writing (2025) · cites this paper as Fleckenstein et al., 2024
Even experts such as professional linguists and teachers are unable to effectively distinguish between human-written and AI-generated text.
“This is heightened by the fact that even experts, such as professional linguists and teachers, are unable to effectively distinguish between human-written and AI-generated text (Casal & Kessler, 2023;Fleckenstein et al., 2024).”
From “ <scp>ChatGPT</scp> can make mistakes. Check important info.” Epistemic beliefs and metacognitive accuracy in students' integration of <scp>ChatGPT</scp> content into academic writing (2025) · cites this paper as Fleckenstein et al., 2024
Cites
ChatGPT for good? On opportunities and challenges of large language models for education
Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., et al. (2023) · Learning and Individual Differences · doi:10.1016/j.lindif.2023.102274 · Cited by 7,326 · see citing works
Chatting and cheating: Ensuring academic integrity in the era of ChatGPT
Cotton, D. R. E., Cotton, P. A., & Shipway, J. R. (2024) · Innovations in Education and Teaching International · doi:10.1080/14703297.2023.2190148 · Cited by 3,087 · see citing works
Chatting and Cheating. Ensuring academic integrity in the era of ChatGPT
Cotton, D., Cotton, P., & Shipway, J. R. (2023) · doi:10.35542/osf.io/mrz8h · Cited by 3,079 · see citing works
Gao, C. A., Howard, F. M., Markov, N. S., Dyer, E. C., Ramesh, S., Luo, Y., & Pearson, A. T. (2022) · doi:10.1101/2022.12.23.521610 · Cited by 1,099 · see citing works
Köbis, N., & Mossink, L. D. (2021) · Computers in Human Behavior · doi:10.1016/j.chb.2020.106553 · Cited by 406 · see citing works
Readers’ perception of computer-generated news: Credibility, expertise, and readability
Graefe, A., Haim, M., Haarmann, B., & Brosius, H.-B. (2018) · Journalism · doi:10.1177/1464884916641269 · Cited by 315 · see citing works
Meyer, J., Jansen, T., Schiller, R., Liebenow, L. W., Steinbach, M., Horbach, A., & Fleckenstein, J. (2024) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2023.100199 · Cited by 311 · see citing works
Self-citationSailer, M., Bauer, E., Hofmann, R., Kiesewetter, J., Glas, J., Gurevych, I., & Fischer, F. (2023) · Learning and Instruction · doi:10.1016/j.learninstruc.2022.101620 · Cited by 106 · see citing works
Keller, S., Fleckenstein, J., Krüger, M., Köller, O., & Rupp, A. (2020) · Journal of Second Language Writing · doi:10.1016/j.jslw.2019.100700 · Cited by 51 · see citing works
Self-citationFleckenstein, J., Keller, S., Krüger, M., Tannenbaum, R. J., & Köller, O. (2020) · Assessing Writing · doi:10.1016/j.asw.2019.100420 · Cited by 50 · see citing works
Self-citationAutomated Essay Scoring at Scale: A Case Study in Switzerland and Germany
Rupp, A., Casabianca, J., Krüger, M., Keller, S., & Köller, O. (2019) · ETS Research Report Series · doi:10.1002/ets2.12249 · Cited by 39 · see citing works
Self-citationJansen, T., Vögelin, C., Machts, N., Keller, S., Köller, O., & Møller, J. (2020) · Teaching and Teacher Education · doi:10.1016/j.tate.2020.103216 · Cited by 37 · see citing works
Self-citationGunser, V. E., Gottschling, S., Brucker, B., Richter, S., Çakir, D., & Gerjets, P. (2022) · IN2WRITING · doi:10.18653/v1/2022.in2writing-1.8 · Cited by 30 · see citing works
Jansen, T., Meyer, J., Fleckenstein, J., Horbach, A., Keller, S., & Møller, J. (2023) · Learning and Instruction · doi:10.1016/j.learninstruc.2023.101847 · Cited by 28 · see citing works
Self-citationJudgment accuracy of German student texts: Do teacher experience and content knowledge matter?
Møller, J., Jansen, T., Fleckenstein, J., Machts, N., Meyer, J., & Reble, R. (2022) · Teaching and Teacher Education · doi:10.1016/j.tate.2022.103879 · Cited by 21 · see citing works
Self-citationEmpathic AI Painter: A Computational Creativity System with Embodied Conversational InteractionPreprint
Yalçın, Ö. N., Abukhodair, N., & DiPaola, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.14223 · Cited by 8 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Human-AI agency in the age of generative AI
Krakowski, S. M. (2025) · Information and Organization · doi:10.1016/j.infoandorg.2025.100560 · Cited by 196 · see citing works
Weng, X.-J., Xia, Q., Gu, M., Rajaram, K., & Chiu, T. (2024) · Australasian Journal of Educational Technology · doi:10.14742/ajet.9540 · Cited by 115 · see citing works
“…thinking, problem-solving and communication skills (Almasre, 2024; Banihashem et al., 2024; Bearman et al., 2024; Blanke, 2024; Cotton et al., 2023; Fleckenstein et al., 2024); and open-ended assessments in which students were encouraged in aspects of originality and creativity (Eager & Brunton,…”
BackgroundInfluentialSimple techniques to bypass GenAI text detectors: implications for inclusive education
Perkins, M., Roe, J., Vu, B. H., Postma, D., Hickerson, D., McGaughran, J., & Khuat, H. Q. (2024) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-024-00487-w · Cited by 117 · see citing works
“This model represented a step-change in the abilities of GenAI to create text that was significantly more difficult to detect by educators (Fleckenstein et al., 2024; Perkins et al., 2023), and thus the technology began to draw greater attention in the public sphere, as well as in educational…”
Acosta-Enriquez, B. G., Várgas, C. G. A. P., Jordan, O. H., Ballesteros, M. A. A., & Morales, A. E. P. (2024) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2024.100320 · Cited by 91 · see citing works
Mizumoto, A., Yasuda, S., & Tamura, Y. (2024) · Applied Corpus Linguistics · doi:10.1016/j.acorp.2024.100106 · Cited by 58 · see citing works
Tejeda, A. S., Oller, J. C. D., Escandell, J. M. B., Gómez-Díaz, R., & Rodríguez, A. G. (2026) · Frontiers in Education · doi:10.3389/feduc.2025.1711718 · Cited by 92 · see citing works
León, C. E. Y., Lipuma, J. M., & Torres, X. O. (2025) · Frontiers in Education · doi:10.3389/feduc.2025.1619888 · Cited by 76 · see citing works
Nguyen, L., & Barrot, J. S. (2024) · Assessing Writing · doi:10.1016/j.asw.2024.100899 · Cited by 67 · see citing works
Rahimi, A. R., & Sevilla‐Pavón, A. (2024) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2024.100258 · Cited by 60 · see citing works
Fiedler, A., & Döpke, J. (2025) · International Review of Economics Education · doi:10.1016/j.iree.2025.100321 · Cited by 40 · see citing works
Heilala, V., Araya, R., & Hämäläinen, R. (2025) · ACM Symposium on Applied Computing · doi:10.1145/3672608.3707764 · Cited by 49 · see citing works
“For example, while LLMs can offer assistance in drafting and editing [48], they also raise questions about academic integrity, as the ease of generating text may compromise the originality of student work [29, 52].”
Use of generative AI in research: ethical considerations and emotional experiences
Farangi, M. R., Nejadghanbar, H., & Hu, G. (2024) · Ethics & Behavior · doi:10.1080/10508422.2024.2420133 · Cited by 45 · see citing works
“This aligns with the literature advocating for human oversight and responsibility in the deployment of AI tools (Fleckenstein et al., 2024; Griffin et al., 2023).”
Urban, M., Brom, C., Lukavský, J., Děchtěrenko, F., Hein, V., Svacha, F., Kmoníčková, P., & Urban, K. (2025) · British Journal of Educational Technology · doi:10.1111/bjet.13591 · Cited by 47 · see citing works
Generative artificial intelligence: implications for biomedical and health professions education
Hersh, W. (2025) · Annual Review of Biomedical Data Science · doi:10.1146/annurev-biodatasci-103123-094756 · Cited by 28 · see citing works
From Superficial Outputs to Superficial Learning: Risks of Large Language Models in EducationPreprint
Delikoura, I., Fung, Y. R., & Hui, P. (2025) · arXiv.org · doi:10.48550/arxiv.2509.21972 · Cited by 28 · see citing works
“Fleckenstein et al. [40] tasked in-service and pre-service teachers with distinguishing between texts produced by ChatGPT and by students.”
Wainaina, P. K., & Sun, Y. (2025) · Discover Education · doi:10.1007/s44217-025-00820-z · Cited by 25 · see citing works
“The sophistication of GenAI intensifies concerns about plagiarism and contract cheating in higher education [9, 16, 25, 56, 65, 69].”
InfluentialFredrick, D. R., & Craven, L. (2025) · Frontiers in Education · doi:10.3389/feduc.2025.1616935 · Cited by 24 · see citing works
Liebenow, L. W., Schmidt, F. T. C., Meyer, J., & Fleckenstein, J. (2025) · Comput. Educ. · doi:10.1016/j.compedu.2025.105385 · Cited by 22 · see citing works
Self-citationAssessing students’ DRIVE: A framework to evaluate learning through interactions with generative AI
Oliveira, M., Zednik, C., Bombaerts, G. J. T., Sadowski, B. M., & Conijn, R. (2025) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2025.100497 · Cited by 23 · see citing works
Kuzu, T. E., Irion, T., & Bay, W. (2025) · Education and Information Technologies · doi:10.1007/s10639-025-13673-8 · Cited by 22 · see citing works
“…systems” but research shows the high potential AI has in supporting processes in professional contexts such as schools (Horbach et al., 2023 ; Meyer et al., 2024), where teachers can use AI to, e.g., solve complex mathematical problems (Kortenkamp & Dohrmann, 2023; Ouyang & Jiao, 2021) but…”
Teaching with ChatGPT as a linguistically responsive tool for multilingual learners
Donley, K. (2024) · Technology in Language Teaching & Learning · doi:10.29140/tltl.v6n3.1719 · Cited by 20
“More specifically related to the context of K–12 language learners, other research has examined how LLM-generated feedback can support secondary students’ writing motivation and positive emotions about their writing, which are associated with more accurate textual revisions (Meyer et al., 2024).”
Perception of ChatGPT Usage for Homework Assignments: Students’ and Professors’ Perspectives
Krečar, I. M., Kolega, M., & Jurcec, L. (2024) · IAFOR Journal of Education · doi:10.22492/ije.12.2.02 · Cited by 20 · see citing works
“…can produce high-quality materials in different disciplines (Herbold et al., 2023; Susnjak, 2022; Yeadon et al., 2023) and the educators can hardly distinguish students’ work from the work written by ChatGPT (Busch & Hausvik, 2023; Fleckenstein et al., 2024; Liu et al., 2023; Waltzer et al., 2023).”
BackgroundResponsible and Ethical Use of AI in Education: Are We Forcing a Square Peg into a Round Hole?
Amigud, A., & Pell, D. J. (2025) · World · doi:10.3390/world6020081 · Cited by 21 · see citing works
“[19] found that “. . . with relatively little prompting, current AI can generate texts that are not detectable for teachers, which poses a challenge to schools and universities in grading student essays”.”
Aluko, H. A., Aluko, A., Offiah, G. A., Ogunjimi, F., Aluko, A. O., Alalade, F. M., Ukeje, I. O., & Nwani, C. H. (2025) · International journal of organizational analysis · doi:10.1108/ijoa-07-2024-4632 · Cited by 20 · see citing works
Sun, Y., Liao, Y.-H., & Ma, X.-X. (2026) · Comput. Educ. · doi:10.1016/j.compedu.2026.105616 · Cited by 17 · see citing works
Kirsanov, O., Kushwah, L., & Selvaretnam, G. (2025) · Journal of Academic Ethics · doi:10.1007/s10805-025-09691-3 · Cited by 16 · see citing works
“…studies on disclosure, motivation, and institutional ambiguity in the wake of generative AI adoption in higher education (e.g. Fawns et al., 2024; Fleckenstein et al., 2024; Freeman, 2025; Stone, 2024), but also adds a behavioural dimension by tracking early opt-out patterns and collecting…”
The AI Memory Gap: Users Misremember What They Created With AI or Without
Zindulka, T., Goller, S., Fernandes, D., Welsch, R., & Buschek, D. (2025) · International Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791494 · Cited by 15
“Thus, here we relate our work to research on ideation with AI and writing with AI.”
Divekar, R. R., Gonzalez, L., Guerra, S., Boos, N., & Divekar, R. (2025) · TechTrends · doi:10.1007/s11528-025-01095-9 · Cited by 20 · see citing works
“…appropriate proctoring may be necessary in some contexts, we advise against relying on AI detection software, as research (Casal & Kessler, 2023; Fleckenstein et al., 2024; Wu et al., 2025) consistently demonstrates that neither humans nor machines can reliably distinguish between AI-generated…”
Participant Use of Artificial Intelligence in Online Focus Groups: An Experiential Account
Stafford, L., Preston, C., & Pike, A. C. (2024) · International Journal of Qualitative Methods · doi:10.1177/16094069241286417 · Cited by 18 · see citing works
“Gibson and Beattie (2024) and Fleckenstein et al. (2024) have noted a lack of mistakes (or ‘ typos ’ ) as an indicator of an LLM response.”
Exploring the Impact of Artificial Intelligence on Research Ethics - A Systematic Review
Andrade-Hidalgo, G., Mio-Cango, P., & Iparraguirre-Villanueva, O. (2024) · Journal of Academic Ethics · doi:10.1007/s10805-024-09579-8 · Cited by 17 · see citing works
“However, its ability to reference the literature is still far from replicating human judgement (Fleckenstein et al., 2024).”
Prompting for creative problem-solving: A process-mining study
Urban, M., Lukavský, J., Brom, C., Hein, V., Svacha, F., Děchtěrenko, F., & Urban, K. (2025) · Learning and Instruction · doi:10.1016/j.learninstruc.2025.102156 · Cited by 15 · see citing works
Assessing EFL students’ GenAI-assisted writing: Teachers’ pains, perceptions and practices
Chen, X., & Pi, X. (2025) · Assessing Writing · doi:10.1016/j.asw.2025.100996 · Cited by 12 · see citing works
The race for AI skills as an obstacle course: Institutional challenges and low threshold suggestions
Vettori, O., & Warm, J. (2025) · Project Leadership and Society · doi:10.1016/j.plas.2025.100183 · Cited by 15 · see citing works
Pretorius, M., & Thewissen, J. (2026) · Journal of Second Language Writing · doi:10.1016/j.jslw.2026.101286 · Cited by 15 · see citing works
Choosing Between an LLM versus Search for Learning: A HigherEd Student PerspectivePreprint
Divekar, R. R., Guerra, S., Gonzalez, L., & Boos, N. (2024) · arXiv.org · doi:10.48550/arxiv.2409.13051 · Cited by 10 · see citing works
“However, research suggests there is no reliable way to detect AI-generated content either by humans [5, 15] ormachines [43].”
Background“Can (A)I do this task?” The role of AI as a socializer of students' self-beliefs of their abilities
Jansen, T., Meyer, J., Fleckenstein, J., Wigfield, A., & Møller, J. (2025) · Learning and Individual Differences · doi:10.1016/j.lindif.2025.102731 · Cited by 12 · see citing works
Self-citationAsamoah, P., Zokpe, D., Boateng, R., Marfo, J. S., Boateng, S. L., Asamoah, D., Muntaka, A. S., & Manso, J. F. (2024) · Cogent Education · doi:10.1080/2331186x.2024.2439651 · Cited by 16 · see citing works
“GenAI and its impact on education has received substantial scholarly attention recently (Saleem et al., 2024; Gr (cid:1) ajeda et al., 2024; Pradana et al., 2023; Qian et al., 2023; Chiu, 2024; Fleckenstein et al., 2024).”
The role of epistemic beliefs in predicting ChatGPT adoption and avoidance in higher education
Urhahne, D., Fink, L., Dietrich, L., & Kremer, K. (2026) · Acta Psychologica · doi:10.1016/j.actpsy.2026.106334 · Cited by 9 · see citing works
Khojasteh, L., Shokrpour, N., & Moslehi, S. (2025) · Teaching and Teacher Education · doi:10.1016/j.tate.2025.104952 · Cited by 10 · see citing works
AI vs. Human Writing: Developing a Novel Method for Text Authenticity Detection in Education
Department of Computer Science and Engineering, R. I. O. T., Kalmani, V. H., Adamuthe, A. C., Premnath Gondil, A., Prashant Patil, V., Amar Kore, R., & Mahadev Metkari, V. (2025) · International Journal of Modern Education and Computer Science · doi:10.5815/ijmecs.2025.03.04 · Cited by 7 · see citing works
School leaders’ expectations of AI’s impact on students and teachers: Insights from ICILS 2023
Eryılmaz, N., Mah, D.-K., Bellibaş, M. Ş., & Pietsch, M. (2025) · Educational Management Administration & Leadership · doi:10.1177/17411432251401487 · Cited by 9 · see citing works
Shafiqa, A. F., Marnaoui, S., & Anqodi, H. A. A. (2024) · Arab World English Journal · doi:10.24093/awej/vol15no4.2 · Cited by 7
“Fleckenstein et al. (2024) point out that current AI tools can generate texts that teachers cannot identify as AI-generated, posing a challenge for schools and universities when grading students’ essays.”
(2025) · Journal of Applied Learning & Teaching · doi:10.37074/jalt.2025.8.1.20 · Cited by 7 · see citing works
Validating automated assessments of teaching effectiveness using multimodal data
Fütterer, T., Hou, R., Bühler, B., Bozkir, E., Bell, C. A., Kasneci, E., Gerjets, P. H., & Trautwein, U. (2025) · Learning and Instruction · doi:10.1016/j.learninstruc.2025.102264 · Cited by 5 · see citing works
Limits of LLM Text Detectors in EducationPreprint
Gehring, L., & Paaßen, B. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.08096 · Cited by 6 · see citing works
“However, manual detection remains unreliable even for experienced teachers [7, 17].”
Lima, A. J. S., Senados, J. M., Senturias, M. G. R., & Simagala, M. J. L. (2025) · International Conference on Educational and Information Technology · doi:10.1109/iceit64364.2025.10975974 · Cited by 9 · see citing works
“Such a high level of knowledge reflects a generally positive outlook on their preparedness and readiness to effectively integrate and utilize ChatGPT in their future teaching practices [10, 12].”
On perception of prevalence of cheating and usage of generative AI
Denkin, R. (2024) · arXiv · doi:10.48550/arxiv.2405.18889 · Cited by 5 · see citing works
“Their findings reveal that both novice and experienced teachers struggle to identify texts generated by ChatGPT, highlighting the difficulties in maintaining academic integrity with the rise of Generative AI [1].”
BackgroundTitova, S. V., & Temuryan, C. T. (2025) · Вестник Тамбовского университета. Серия: Гуманитарные науки · doi:10.20310/1810-0201-2025-30-5-1038-1062 · Cited by 5 · see citing works
“Эти дан-ные можно объяснить тем, что на сегодняш-ний момент проверка ИИ несовершенна и требует контроля и перепроверки педагогом [30], ИИ достаточно редко используется для языкового контроля, исключение составляет формирующее оценивание [31].”
Reconstructing stance in EFL doctoral thesis writing through generative artificial intelligence
Zhao, W. (2025) · Humanities and Social Sciences Communications · doi:10.1057/s41599-025-06249-x · Cited by 8 · see citing works
“Existing research has primarily focused on areas such as Generative AI writing assessment (e.g. Guo and Wang 2023, Zhang and Hyland 2018), teachers ’ use of Generative AI (e.g. Fleckenstein et al. 2024, Zhao 2024), and Generative AI-generated feedback (e.g. Kasneci et al. 2023, Lund et al. 2023).”
Detecting LLM-Generated Short Answers and Effects on Learner Performance
Bhushan, S., Thomas, D. R., Borchers, C., Raghuvanshi, I., Abboud, R., Gatz, E., Gupta, S., & Koedinger, K. (2025) · EC-TE · doi:10.48550/arxiv.2506.17196 · Cited by 5
“The study relies on human judgment as the ground truth, which may not always be correct and introduce bias or inconsistencies [14].”
Ly, R., & Ly, B. (2025) · AI and Ethics · doi:10.1007/s43681-025-00667-y · Cited by 6 · see citing works
Ethical procedures for responsible experimental evaluation of AI-based education interventions
Dekker, I., Bert Bredeweg, A., Winkel, W. T., & van de Poel, I. (2024) · AI and Ethics · doi:10.1007/s43681-024-00621-4 · Cited by 8 · see citing works
“Furthermore, it could be used not only for feedback but for first or even final drafts of essays, which might lead the professional to reconsider how to grade essays and or use them for assessment [e.g., 15].”
Ellison, C., Coulby, C., Saunders, S., Lindsay, R., & Hands, C. (2026) · Perspectives: Policy and Practice in Higher Education · doi:10.1080/13603108.2025.2505883 · Cited by 5 · see citing works
“To date this has been shown to be impossible for both algorithms (Odri and Yoon 2023) and people (Fleckenstein et al. 2024).”
Albaloul, O., Alajmi, A., Oster, N. J., & Killian, C. M. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-01344-7 · Cited by 3 · see citing works
Cespedes, A. A., & Tursunkhanova, E. (2026) · Journal of Academic Ethics · doi:10.1007/s10805-025-09718-9 · Cited by 4 · see citing works
“Scholars even suggest that some teachers are over confident in identifying supposedly AI-written student work (Fleckenstein et al., 2024 ), which can lead to false accusations.”
A Multimodal Framework for Explainable Evaluation of Soft Skills in Educational Environments
Guerrero-Sosa, J. D., Romero, F. P., Menéndez-Domínguez, V., Serrano-Guerrero, J., Montoro-Montarroso, A., & Olivas, J. (2025) · arXiv.org · doi:10.48550/arxiv.2505.01794 · Cited by 4
“The advent of LLMs introduces several complexities in assessing skills [18] such as clarity, critical thinking and expression because these models may bypass traditional learning processes [19].”
Gupta, V., & Gupta, C. (2024) · Computer · doi:10.1109/mc.2024.3445068 · Cited by 4 · see citing works
Funa, A. A., & Gabay, R. A. E. (2025) · Social Sciences & Humanities Open · doi:10.1016/j.ssaho.2025.102358 · Cited by 9 · see citing works
Assessment validity in the age of generative artificial intelligence: a critical review
Ali, R., & Maroulis, J. (2026) · Assessment & Evaluation in Higher Education · doi:10.1080/02602938.2026.2683046 · Cited by 3 · see citing works
The Impact of AI on School Leadership Practices and Policy Enactment: A Scoping Literature Review
Mifsud, D., Campbell, P., & Niesche, R. (2026) · Leadership and Policy in Schools · doi:10.1080/15700763.2026.2679510 · Cited by 6 · see citing works
Artificial Intelligence in Language Assessment Research: A Scoping Review
O’Grady, S., Göktürk, N., & Rossi, O. (2026) · Research Synthesis in Applied Linguistics · doi:10.1080/29984475.2026.2633389 · Cited by 8 · see citing works
Ai-RACE as a Framework for Writing Assignment Design in Higher Education
El-Soussi, A., & Yousef, D. (2026) · Education Sciences · doi:10.3390/educsci16010119 · Cited by 4 · see citing works
Oladokun, B. D., Ukubeyinje, S. E., Oyighan, D., Obim, I. E., & Oloniruha, E. A. (2025) · College & Undergraduate Libraries · doi:10.1080/10691316.2025.2607372 · Cited by 3 · see citing works
“These difficulties are compounded by the evolving nature of AI-generated content which can seamlessly mimic human writing styles, making detection increasingly difficult (Fleckenstein et al., 2024; Perkins et al., 2024).”
InfluentialLow, D. Y. S., Shaw, K. M., & Balan, P. (2025) · Sustainable Development · doi:10.1002/sd.70446 · Cited by 3 · see citing works
Learning to detect AI texts and learning the limits
Milička, J., Marklová, A., Drobil, O., & Pospíšilová, E. (2025) · PLoS ONE · doi:10.1371/journal.pone.0333007 · Cited by 3 · see citing works
BENARAB, I. H. (2024) · أطراس · doi:10.70091/atras/ai.17 · Cited by 3 · see citing works
“In this context, and to assess teachers' ability to identify texts generated by AI, in particular those from OpenAI's ChatGPT, Fleckenstein et al. (2024) conducted a study with experiments involving 89 novice and 200 experienced teachers.”
MethodElkhodr, M., & Gide, E. (2026) · STEM Education · doi:10.3934/steme.2026024 · Cited by 3 · see citing works
Reinstituting teacher agency and classroom resilience amid geopolitics of AI homogenisation
Ficapal‐Cusí, P., Rajmil, D., & Torrent‐Sellens, J. (2026) · Pedagogy Culture and Society · doi:10.1080/14681366.2026.2709075 · Cited by 2 · see citing works
Akpınar, Y., & Kutbay, E. (2026) · The Turkish Online Journal of Distance Education · doi:10.17718/tojde.1705032 · Cited by 2 · see citing works
Menze, D., Radović, S., & Seidel, N. (2026) · Frontiers of Computer Science · doi:10.3389/fcomp.2026.1831250 · Cited by 2 · see citing works
Cui, Y. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-01286-0 · Cited by 9 · see citing works
“However, AI’s limitations in evaluating complex academic writing, particularly in identifying implicit relationships and critical thinking depth, remain a significant challenge [15, 24].”
Mohamed, R. A. A. (2026) · Interactive Learning Environments · doi:10.1080/10494820.2026.2649549 · Cited by 3 · see citing works
“Hallucinations and unwarranted confidence can foster automation bias (Bender et al., 2021; Ji et al., 2023; Lee & See, 2004), and easy text generation raises integrity and authorship concerns (Draxler et al., 2023; Fleckenstein et al., 2024).”
Assessing L2 Student Writing in the AI Era: A Systematic Review on Challenges and Best Practices
Dau, T. T. L., & Lam, T. T. (2026) · International Journal of TESOL & Education · doi:10.54855/ijte.26615 · Cited by 2 · see citing works
“…widely recommended AI literacy training for teachers, especially on how to identify features of AI-created content (Alexander et al., 2023; Fleckenstein et al., 2024), and teachers themselves need to do more practice to get their grading fair and accurate in students’ assignments (Waltzer…”
InfluentialFiedler, D., Schönle, D., Reich, C., & Harms, U. (2025) · Journal of Computer Assisted Learning · doi:10.1002/jcal.70161 · Cited by 4 · see citing works
Navigating nursing student use and suspected misuse of artificial intelligence
Chicca, J., & Shellenbarger, T. (2025) · Teaching and Learning in Nursing · doi:10.1016/j.teln.2025.11.020 · Cited by 2 · see citing works
Fitriyah, F., & Ratnawati, R. (2025) · Jurnal Wahana Pendidikan · doi:10.25157/jwp.v12i2.19776 · Cited by 2
“Teachers are having more and more trouble identifying AI-generated content, which makes it harder to evaluate students' unique work and compromises the validity of performance reviews (Fleckenstein et al., 2024).”
Choi, J., Zhao, J., Ngo, T.-A., & Grierson, L. (2025) · Canadian Medical Education Journal · doi:10.36834/cmej.79690 · Cited by 2 · see citing works
“14 22 Numerous studies highlight how AI generates believable high-quality texts that can improve upon or outperform human writing, 15–18 and that can replicate the type of empathy we desire in physicians.”
Challenges and Opportunities of Using Generative AI for Research: Opening the Discussion
Goretti, L., Küsters, A., Morgana, S., & Sharma, A. (2024) · The International Spectator · doi:10.1080/03932729.2024.2435176 · Cited by 3 · see citing works
“For example, experiments have shown that generative AI can produce essays that are indistinguishable from those written by students, making it di ffi cult for educators to detect plagiarism (Fleckenstein et al. 2024).”
From Data to Decisions: Leveraging AI for Proactive Education Strategies
Moore, W., & Tsay, L.-S. (2024) · International Conference on AI Research · doi:10.34190/icair.4.1.3082 · Cited by 2
Selvam, V. T., & Zakaria, N. Y. K. (2024) · International Journal of Academic Research in Progressive Education and Development · doi:10.6007/ijarped/v13-i3/22863 · Cited by 2
“Fleckenstein et al. (2024) on the other hand, examine teachers' ability to detect AI-generated texts, stressing the need for educators to adapt their assessment methods to distinguish between human and AI-generated work to maintain academic integrity.”
Can GPTZero's AI Vocabulary Distinguish Between LLM-Generated and Student-Written Essays?
Schmalz, V., & Tack, A. (2025) · Workshop on Innovative Use of NLP for Building Educational Applications · doi:10.18653/v1/2025.bea-1.71 · Cited by 2
“Even experienced teachers and those familiar with LLMs often struggle to tell apart student-written essays from those created by LLMs, as studies have shown (Fleckenstein et al., 2024; Waltzer et al., 2024; Perkins et al., 2024).”
Leveling the Playing Field? Generative AI and Entrepreneurial Storytelling
Taeuscher, K., & Lounsbury, M. (2026) · Entrepreneurship: Theory & Practice · doi:10.1177/10422587261472523 · Cited by 1 · see citing works
Mifsud, D., Campbell, P., & Niesche, R. (2026) · Management in Education · doi:10.1177/08920206261475462 · Cited by 1 · see citing works
“Concerns about academic integrity, overreliance on automation and diminished creativity remain prevalent (Fleckenstein et al., 2024).”
Kumar, J. A., Salim, Z., Annamalai, N., Dhaliwal, J., & Singh, A. D. (2026) · Quality & Quantity · doi:10.1007/s11135-026-02982-w · Cited by 1 · see citing works
Contradictions and Cues: Understanding Educators’ Judgments of GenAI-Assisted Essays
Findon, J. L., Upsher, R., Heard, C., Yalcintas, S., Cotier, F. A., & Runswick, O. R. (2026) · Innovación educativa · doi:10.1163/25248502-bja00017 · Cited by 1 · see citing works
Integrating generative AI in first-year writing: Lessons from a pilot initiative
McVey, C., Fassihi, P., Geng, S., Hardy, S., & Parfitt, M. (2026) · Computers and Composition · doi:10.1016/j.compcom.2026.102982 · Cited by 1 · see citing works
Mizumoto, A. (2026) · Applied Corpus Linguistics · doi:10.1016/j.acorp.2026.100223 · Cited by 1 · see citing works
Beyond Accuracy: Experts See AI Fact-Checks as Accurate but Less Useful
Jia, C., Gondimalla, A., Zhang, A., Mullings, D. J., Boltz, A., & Lee, M. K. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791625 · Cited by 1 · see citing works
Zou, Y., Kuek, F., Ng, K. H., & Cheng, X. (2026) · PLoS ONE · doi:10.1371/journal.pone.0343163 · Cited by 3 · see citing works
“Tools like ChatGPT and Kimi can generate coherent, contextually relevant texts in diverse fields [7], enhancing productivity [1,8] and at times, mimicking human writing convincingly [9,10].”
AL-Rawas, M., Qader, O. A. J. A., Lin, G. S. S., Beh, Y. H., Sabarudin, M. A., Ang, Y., Low, J. F., Abdullah, J. Y., & Noorani, T. Y. (2026) · Scientific Reports · doi:10.1038/s41598-026-42555-3 · Cited by 3 · see citing works
“One study demonstrated humans can correctly identify AI-generated text approximately 53% of the time 17 .”
InfluentialLiu, S., Li, X., & Hao, J. (2026) · Interactive Learning Environments · doi:10.1080/10494820.2026.2635540 · Cited by 1 · see citing works
“…researches have reported that GenAI is already widely used in higher education for learning support, content generation, and formative feedback (Lee & Palmer, 2025), which cause the teachers difficulty in discerning whether the answers are given by human or GenAI (Fleckenstein et al., 2024).”
InfluentialMethodological Variation in Studying Staff and Student Perceptions of AI
Gerard, J., Macleod, M., Norwood, K., Reid, A., & Singh, M. (2025) · Frontiers in Education · doi:10.48550/arxiv.2602.11158 · Cited by 1
“There is consensus that new strategies are needed to handle this — whether through assessment design, honour codes, or requiring students to show their AI inputs and editing process, aligning with recent research on detection rates (Fleckenstein et al., 2024; Perkins et al., 2024).”
Generative AI: A Disruptor to Health Professions Learner Assessment
Morjaria, L., Gandhi, B., Burns, L., & Sibbald, M. (2025) · Cureus · doi:10.7759/cureus.96034 · Cited by 1 · see citing works
“This has made it difficult for assessors to accurately and reliably identify text written by students or AI [13].”
Abilities of design professors to distinguish design assignments generated by students and AI
Yin, Y., Wang, B.-H., Zuo, H., Liu, R., Vohra, S., Haydon-Rowe, S., & Childs, P. R. N. (2025) · Proceedings of the Design Society · doi:10.1017/pds.2025.10046 · Cited by 1
“Fleckenstein et al. (2024) asked young and professional educators to assess the writing generated by students and ChatGPT.”
English Teachers’ Experiences with Artificial Intelligence-Driven Students’ Outputs in the Classroom
Bandao, R. B. (2025) · International Journal of Innovative Science and Research Technology · doi:10.38124/ijisrt/25jul299 · Cited by 1 · see citing works
Divekar, R. R., Guerra, S., Gonzalez, L., Boos, N., & Zhou, H. (2025) · International Conference on Artificial Intelligence in Education · doi:10.1007/978-3-031-98465-5_48 · Cited by 1 · see citing works
“However, the new homework machine contextualizes the information, so it can fit the rubric of a unique homework assignment, masquerading as human written since plagiarism is difficult to detect by both humans [2] and machines [3].”
Student or AI? Automated Detection of AI-generated Student Essays
Boutadjine, A., Harrag, F., & Shaalan, K. F. (2026) · Procedia Computer Science · doi:10.1016/j.procs.2026.01.112 · Cited by 1 · see citing works
GİRAY, L., CANER, H. N., & AMOOZEGAR, A. (2026) · Journal of Narrative and Language Studies · doi:10.59045/nalans.2026.109
Baldus, B. J. (2026) · Journal of Marketing Education · doi:10.1177/02734753261491247
Silang, C. D. (2026) · QualiSearch Journal of Educational Research and Practice · doi:10.66437/qjerp.v2si.017
- Authors
- Fleckenstein, J., Meyer, J., Jansen, T., Keller, S. D., Köller, O., & Möller, J.
- Affiliations
- University of Hildesheim (DE)
- Leibniz Institute for Science and Mathematics Education (DE)
- Zurich University of Teacher Education (CH)
- Corresponding author
- Fleckenstein, Johanna
- Venue
- Computers and Education: Artificial Intelligence
- Publisher
- Elsevier BV
- Published
- 2024-06
- Volume
- 6
- Pages
- 100209
- ISSN
- 2666-920X (ISSN-L)
- OpenAlex
- W4391222332
- Cited by
- 126
- References
- 16
- Citations by year
- 2026: 58 · 2025: 48 · 2024: 20
- FWCI
- 47.38
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Computers and Education Artificial Intelligence (publishedVersion, cc-by-nc-nd)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
- Social Sciences › Safety Research › Academic integrity and plagiarism
- Physical Sciences › Computer Science › Artificial Intelligence › Authorship Attribution and Profiling
- Keywords
- AI-generated text detection, generative AI, ChatGPT, student essays, teacher judgment, essay scoring, overconfidence, source effects, academic integrity
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
