Students are using large language models and AI detectors can often detect their use
Open accessAbstract
Large language model (LLM) artificial intelligence (AI) has been in development for many years. Open AI thrust them into the spotlight in late 2022 when it released ChatGPT to the public. The wide availability of LLMs resulted in various reactions, from jubilance to fear. In academia, the potential for LLM abuse in written assignments was immediately recognized, with some instructors fearing they would have to eliminate this mode of evaluation. In this study, we seek to answer two questions. First, how are students using LLM in their college work? Second, how well do AI detectors function in the detection of AI-generated text? We organized 153 students from an introductory microbiology course to write essays on the regulation of the tryptophan operon. We then asked AI the same question and had the students try to disguise the answer. We also surveyed students about their use of LLMs. The survey found that 46.9% of students use LLM in their college work, but only 11.6% use it more than once a week. Students are unclear about what constitutes unethical use of LLMs. Unethical use of LLMs is a problem, with 39% of students admitting to using LLMs to answer assessments and 7% using them to write entire papers. We also tested their prose against five AI detectors. Overall, AI detectors could differentiate between human and AI-written text, identifying 88% correctly. Given the stakes, having a 12% error rate indicates we cannot rely on AI detectors alone to check LLM use, but they may still have value.
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?
These limitations are frequently framed in terms of the risk of false positives, which may in turn lead to false accusations of academic misconduct.
“These limitations are frequently framed in terms of the risk of false positives (Paustian & Slinger, 2024), which may in turn leads to false accusations of academic misconduct.”
From Assessment-focused governance mode? Mapping Generative AI policies across 76 UK higher education institutions: Policy framework and recommendations (2026) · cites this paper as Paustian & Slinger, 2024
AI-detection software programs are currently insufficiently accurate at identifying AI-generated text to be used as a sole criterion for detection.
“However, as described previously, such programs are currently insufficiently accurate at identifying AI-generated text to be used as a sole criterion for detection.”
From Maintaining Academic Integrity Through AI-Trap Questions (2025) · cites this paper as Paustian & Slinger, 2024
AI detection programs had an 88% accuracy rate.
“A separate team of researchers found that student use of AI was prevalent and that AI detection programs had an 88% accuracy rate (Paustian & Slinger, 2024).”
From Maintaining Academic Integrity Through AI-Trap Questions (2025) · cites this paper as Paustian & Slinger, 2024
An 88% accuracy rate was insufficient to be relied upon, at least exclusively.
“They concluded that such a rate was insufficient to be relied upon, at least exclusively.”
From Maintaining Academic Integrity Through AI-Trap Questions (2025) · cites this paper as Paustian & Slinger, 2024
Student use of AI was prevalent.
“A separate team of researchers found that student use of AI was prevalent and that AI detection programs had an 88% accuracy rate (Paustian & Slinger, 2024).”
From Maintaining Academic Integrity Through AI-Trap Questions (2025) · cites this paper as Paustian & Slinger, 2024
Cites
GPT detectors are biased against non-native English writers
Liang, W., Yuksekgonul, M., Mao, Y., Wu, E., & Zou, J. (2023) · Patterns · doi:10.1016/j.patter.2023.100779 · Cited by 673 · 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
Elkhatat, A. M., Elsaid, K., & Almeer, S. (2023) · International Journal for Educational Integrity · doi:10.1007/s40979-023-00140-5 · Cited by 392 · see citing works
Formative Assessment and Writing: A Meta-Analysis
Graham, S., Hebert, M., & Harris, K. R. (2015) · The Elementary School Journal · doi:10.1086/681947 · Cited by 340 · see citing works
All That’s ‘Human’ Is Not Gold: Evaluating Human Evaluation of Generated Text
Clark, E., August, T., Serrano, S., Haduong, N., Gururangan, S., & Smith, N. A. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.565 · Cited by 290 · see citing works
All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Text
Clark, E., August, T., Serrano, S., Haduong, N., Gururangan, S., & Smith, N. A. (2021) · arXiv · doi:10.48550/arxiv.2107.00061 · Cited by 290
Why do students cheat? Perceptions, evaluations, and motivations
Waltzer, T., & Dahl, A. (2022) · Ethics & Behavior · doi:10.1080/10508422.2022.2026775 · Cited by 87 · see citing works
Assisting academics to identify computer generated writing
Abd-Elaal, E.-S., Gamage, S. H. P. W., & Mills, J. E. (2022) · European Journal of Engineering Education · doi:10.1080/03043797.2022.2046709 · Cited by 62 · see citing works
Cognitive Mirage: A Review of Hallucinations in Large Language Models
Ye, H., Liu, T., Zhang, A., Hua, W., & Jia, W. (2023) · arXiv · doi:10.48550/arxiv.2309.06794 · Cited by 61 · see citing works
Gunser, V. E., Gottschling, S., Brucker, B., Richter, S., & Gerjets, P. (2021) · Communications in Computer and Information Science · doi:10.1007/978-3-030-78635-9_67 · Cited by 25 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Tiller, N. B., Marcon, A. R., Zenone, M. A., Kidd, K. E., Jeukendrup, A. E., Master, Z., & Caulfield, T. (2026) · BMJ Open · doi:10.1136/bmjopen-2025-112695 · Cited by 43 · see citing works
Exploring the Impact of LLM Prompting on Students’ Learning
Mutanga, M. B., Msane, J., Mndaweni, T. N., Hlongwane, B. B., & Ngcobo, N. Z. (2025) · Trends in Higher Education · doi:10.3390/higheredu4030031 · Cited by 23 · see citing works
Giray, L., Jacob, J., Encanto, V., & Mansilungan, C. J. (2025) · Journal of Academic Ethics · doi:10.1007/s10805-025-09695-z · Cited by 20 · see citing works
Overviewing the Maze of Research Integrity and False Positives Within AI-Enabled Detectors
Abubakari, M. S. (2024) · Advances in educational technologies and instructional design book series · doi:10.4018/979-8-3693-0884-4.ch013 · Cited by 12 · see citing works
Yeh, S. N., & Siah, C. J. R. (2025) · Nurse Education in Practice · doi:10.1016/j.nepr.2025.104567 · Cited by 11 · see citing works
Klemettilä, P. A., Sharma, S., Mochiyama, F., Iivari, N., Iwata, M., & Koivisto, J. (2025) · ACM International Conference on Interaction Design and Children (IDC) · doi:10.1145/3713043.3728853 · Cited by 6 · see citing works
Freitas, G., Ünal, I., Goldust, M., & Grant-Kels, J. M. (2026) · JAAD reviews. · doi:10.1016/j.jdrv.2026.03.003 · Cited by 5 · see citing works
Factors influencing faculty decision-making on student use of generative AI in higher education
Dilinika, J. M. S. (2026) · Journal of Interdisciplinary Studies in Education · doi:10.32674/ksb6gv95 · Cited by 5 · see citing works
Maintaining Academic Integrity Through AI-Trap Questions
Stillman, T. (2025) · Journal of Educators Online · doi:10.9743/jeo.2025.22.4.14 · Cited by 2 · see citing works
Designing Education for the AI Era: Principles for Integrating LLMs in Pedagogy
Biscalchin, A., Sarkheyli-Hägele, A., Eriksson, J., & Vogel, B. (2025) · International Conference on Software, Telecommunications and Computer Networks (SoftCOM) · doi:10.23919/softcom66362.2025.11197458 · Cited by 2 · see citing works
A Survey on AI-Generated Text Detection: Methods, Challenges, and Future Directions
Ismail, A. A., Ahmed, C., & ElKorany, A. (2025) · Future International Conference on Artificial Intelligence and Cybersecurity (FICAC) · doi:10.1109/ficac65757.2025.11341850 · Cited by 1 · see citing works
Boundary Conditions for AU-Based Detection of Understanding: A Literary Analysis Study
Lazic, M., & Woodruff, E. (2026) · Electronics · doi:10.3390/electronics15102059 · Cited by 1 · see citing works
Generative Artificial Intelligence and English-Medium Education
Moorhouse, B. L., Pun, J., & Curle, S. M. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009736862 · Cited by 1 · see citing works
Balas, M., Li, X., Ji, P., & Arjmand, P. (2025) · Canadian Journal of Ophthalmology · doi:10.1016/j.jcjo.2025.09.004 · Cited by 1 · see citing works
Jelica, S., Milenković, I., Minović, M., & Maričić, M. (2025) · Computer Applications in Engineering Education · doi:10.1002/cae.70105 · Cited by 1 · see citing works
Bruno: A RAG-based Programming Learning Assistant for Guided, No-Code Problem Solving
Daugdaug, J. M., Salera, L. V. A., Alosada, H. G., Busico, E. C., & Amparo, J. C. (2025) · International Conference on Education and E-Learning (ICEEL) · doi:10.1145/3789859.3789864 · Cited by 1 · see citing works
Bostancioglu, A., Daş, B., & Özek, M. B. (2026) · Applied Sciences · doi:10.3390/app16147249 · Cited by 1 · see citing works
Oral exams in introductory statistics class with non‐native English speakers
Yanchenko, E. (2025) · Teaching Statistics · doi:10.1111/test.70015 · Cited by 1 · see citing works
Tao, Y., Ryan, M., Bialy, F., & Zhang, X. (2026) · Research Square (preprint) · doi:10.21203/rs.3.rs-10029667/v1
Coriell, J., Kiremire, A. R., Cruse, K., & Long, W. (2025) · ASEE Annual Conference & Exposition · doi:10.18260/1-2--55907
The Application of Artificial Intelligence in University Teaching
Wang, Y. (2025) · Lecture Notes in Education Psychology and Public Media · doi:10.54254/2753-7048/2025.ns26559
Comparing Physics Examination Performance with and without ChatGPT Assistance
Radenković, L., Nešić, L., & Radovančević, D. (2025) · The Physics Teacher · doi:10.1119/5.0240794
Alhokail, L. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-3451-6.ch007
AI See What You Did There – The Prevalence of LLM-Generated Answers in MOOC Responses
Nurmi, P., Khan, M., Safaei, Z., Nguyen, N. T., Sarhaddi, F., Tompuri, M., Nygren, H., Kinnunen, P., & Zuniga, A. (2026) · ACM Technical Symposium on Computer Science Education (SIGCSE) · doi:10.1145/3770762.3772602
Usage Behaviors of Large Language Models: Taking Undergraduate Students as an Example
Chen, C.-C., & Lu, H.-Y. (2026) · Journal of Advanced Computational Intelligence and Intelligent Informatics · doi:10.20965/jaciii.2026.p0343
Psikologi Idiolek: Erosi Kepemilikan Linguistik dalam Praktik Menulis dengan AI
Samosir, Z. C., Marpaung, Y. D. M., Syafira, M. D., Azzahra, M. A., & Adi, P. (2026) · Jurnal Pustaka Cendekia Hukum dan Ilmu Sosial · doi:10.70292/pchukumsosial.v4i1.289
Etika Akademik Penggunaan AI Dalam Penulisan Tugas Bagi Mahasiswa Sejarah
Afida, S. P., & Purwaningsih, S. M. (2026) · EFEKTOR · doi:10.29407/e.v13i1.28705
Acoustic Correlates of Phases of Understanding in a Naturalistic Literary Task
Lazic, M., & Woodruff, E. (2026) · Electronics · doi:10.3390/electronics15122700
Confidence Without Competence in AI-Assisted Knowledge Work
Eleftheriou, E., Pallis, G., & Constantinides, M. (2026) · Annual Symposium on Human-Computer Interaction for Work (CHIWORK) · doi:10.1145/3808045.3808055
Zhou, K., Littlejohn, M., Garrard, I., & Barlow, J. (2026) · Journal of Ethics and Emerging Technologies · doi:10.55613/jeet.v36i2.234
AI ASSISTANCE OR OVERRELIANCE? ANALYZING ESP STUDENTS’ CRITICAL USE OF GENERATIVE AI
Ajiza, M., Yuliasri, I., Pratama, H., & Wahyuni, S. (2026) · English Review Journal of English Education · doi:10.25134/englishreview.v14i2.153
Deep, P. D. (2026) · International Journal of Studies in Education and Science · doi:10.46328/ijses.7882
Withanage, W. D. M. D., Nissanka, N. M. T. A., Wickramasinghe, C. P., Fernando, W. P. P. D., Perera, V., & Gunatilake, G. D. (2026) · Access Microbiology · doi:10.1099/acmi.0.001137.v5
Sarmiento, J. B., & Jalani, N. T. (2026) · International Journal of Research on Multidisciplinary Studies · doi:10.67903/ijroms0405
Students’ Perspectives on Generative AI Usage Policies at Universitas Pendidikan Indonesia
Maulana, M., Irawati, R., Mugara, R., Nugraha, B., Adnyana, I. K. W., Ali, E. Y., & Munawaroh, R. (2026) · International Journal of Educational Practice and Policy · doi:10.66314/ijepp.v4i3.1468
- Authors
- Paustian, T., & Slinger, B.
- Affiliations
- University of Wisconsin–Madison (US)
- Corresponding author
- Paustian, Timothy
- ORCID
- Paustian, Timothy
- Venue
- Frontiers in Education
- Publisher
- Frontiers Media SA
- Published
- 2024-06-07
- Volume
- 9
- ISSN
- 2504-284X (ISSN-L)
- OpenAlex
- W4399543927
- Cited by
- 27
- References
- 9
- Citations by year
- 2026: 13 · 2025: 13 · 2024: 1
- FWCI
- 11.24
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Frontiers in Education (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Topics
- Social Sciences › Safety Research › Academic integrity and plagiarism
- Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Keywords
- large language models, AI-generated text detection, student use of AI, academic integrity, ChatGPT, microbiology education, essay writing, detection accuracy
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
