The rapid development of Large Language Models (LLMs) has made it increasingly hard to distinguish human written from AI-generated academic papers, raising questions about authorship, originality, and academic integrity. This research presents a systematic approach to identifying and measuring multidimensional markers that differentiate human written academic manuscripts from AI-generated and AI-paraphrased summaries. After analysing 52 human written academic papers and their AI-generated versions (created with GPT-4o, DeepSeek V4, and Qwen 3.7plus), we developed an automated, multi-layered text-annotation tool. It incorporates rule based matching, zero shot classification, and structured LLMaided annotation, and we validated it using human inter annotator agreement. We found that no single language marker is enough for classification. However, the accurate classification (85-90%) requires a set of markers which include the completeness of the scholarly apparatus (the most reliable one), significant information loss (40-50% of granular details such as exact statistics or software name), decrease of syntactic burstiness (by 40-50%), formulaic transition substitutions, and the drop in epistemic hedging by 60%. Although the absence of structure and metadata yields perfect classification, the stylometric approach alone produced an AUC of 0.78, underscoring the complexity of linguistic signatures. In summary, successfully implementing AI for identifying text in academic environments requires a multi marker approach that favours completeness, informativeness, and text depth over fluency and vocabulary sophistication, which are highly prone to false positive results. Previous research has addressed AI academic detection; however, we shift our emphasis from surface fluency (which LLMs excel at mimicking) to epistemic rigour and granular information loss.
[1] Amirjalili, F., Neysani, M., Nikbakht, A. (2024). Exploring the boundaries of authorship: A comparative analysis of AI-generated text and human academic writing in English literature. Frontiers in Education, 9, 1347421.
[2] Ren, M. (2024). Advancements and applications of large language models in natural language processing: A comprehensive review. Applied and Computational Engineering, 97, 55–63.
[3] Christian, J. (2023, January). Cnet secretly used ai on articles that didn’t disclose that fact, staff say. Futurism. https://futurism.com/cnet-ai-articles-label.
[4] Jurafsky, D., Martin, J. H. (2014). Speech and language processing (Vol. 3). Pearson.
[5] Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D., Dhariwal, P., Amodei, D. (2020). Language models are few-shot learners. Advances in Neural Information Processing Systems, 33, 1877–1901.
[6] Georgiou, G. P. (2026). What distinguishes AI-generated from human writing A rapid review of the literature. Big Data and Cognitive Computing, 10(2), 55. https://doi.org/10.3390/bdcc10020055.
[7] Fariello, S., Fenza, G., Forte, F., Gallo, M., Marotta, M. (2025). Distinguishing human from machine: A review of advances and challenges in AI-generated text detection. International Journal of Interactive Multimedia and Artificial Intelligence, 9(3), 6-18. https://doi.org/10.9781/ijimai.2024.12.002.
[8] Fowler, G. A. (2023, April 1). We tested a new ChatGPT detector for teachers. It flagged an innocent student. The Washington Post. https://www.washingtonpost.com/technology/2023/04/01/chatgpt-cheating-detectionturnitin.
[9] Hill, K. (2022, May 27). Accused of cheating by an algorithm, and a professor she had never met. The New York Times. https://www.nytimes.com/2022/05/27/technology/college-students-cheating-software-honorlock.html.
[10] Das, M. R. (2023). Turn-it-in: Ai fails students for not using ai. Firstpost. https://www.firstpost.com/world/ plagiarism-detector-turnitin-keeps-falsely-accusing-students-of-cheating-using-ai-12704662.html.
[11] Quach, K. (2023, May 17). Professor freezes student grades after ChatGPT claimed AI wrote their papers. The Register. https://www.theregister.com/2023/05/17/university_chatgpt_grades.
[12] Al-Sibai, N. (2023). AI plagiarism detection software keeps falsely accusing students of cheating. Futurism. https://futurism.com/ai-plagiarism-software-false-accusing-students.
[13] Cheng, A., Lin, Y., Reedy, G., Joseph, C., Wirkowski, S., Mallette, V., Calhoun, A. (2025). Ability of AI detection tools and humans to accurately identify different forms of AI-generated written content. Advances in Simulation, 10(1), 66.
[14] Fraser, K. C., Dawkins, H., Kiritchenko, S. (2025). Detecting AI-generated text: Factors influencing detectability with current methods. Journal of Artificial Intelligence Research, 82, 2233-2278.
[15] André, C. M. J., Eriksen, H. F. L., Jakobsen, E. J., Mingolla, L. C. B., Thomsen, N. B. (2023). Detecting AI authorship: Analyzing descriptive features for AI detection. In NL4AI 2023: Seventh Workshop on Natural Language for Artificial Intelligence. CEUR-WS. https://ceur-ws.org/Vol-3551/paper3.pdf.
[16] Liu, Y., Zhang, Z., Zhang, W., Yue, S., Zhao, X., Cheng, X., Zhang, Y., Hu, H. (2023). ArguGPT: Evaluating, understanding and identifying argumentative essays generated by GPT models. arXiv preprint. https://doi.org/ 10.48550/arXiv.2304.07666.
[17] Muñoz-Ortiz, A., Gómez-Rodríguez, C., Vilares, D. (2024). Contrasting linguistic patterns in human and LLM-generated news text. Artificial Intelligence Review, 57(10), 265. https://doi.org/10.1007/s10462-024- 10903-2.
[18] Fedoriv, Y., Pirozhenko, I., Shuhai, A. (2023). Linguistic analysis of human-created and artificial intelligence generated content in academic discourse. Journal of Vasyl Stefanyk Precarpathian National University. Philology, (10), 47–67. https://doi.org/10.15330/jpnuphil.10.47-67.
[19] Terèon, L., Dobrovoljc, K. (2025). Linguistic characteristics of AI-generated text: A survey. arXivpreprint. https://arxiv.org/abs/2510.05136.
[20] Zanotto, S. E., Aroyehun, S. (2025, November). Linguistic and embedding based profiling of texts generated by humans and large language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (pp. 22852-22869).
[21] Alexander, K., Savvidou, C., Alexander, C. (2023). Who wrote this essay Detecting AI-generated writing in second language education in higher education. Teaching English with Technology, 23, 25–43.
[22] Herbold, S., Hautli-Janisz, A., Heuer, U., Kikteva, Z., Trautsch, A. (2023). A large-scale comparison of human written versus ChatGPT-generated essays. Scientific Reports, 13, 18617.
[23] Cai, Z. G., Duan, X., Haslett, D. A., Wang, S., Pickering, M. J. (2023). Do large language models resemble humans in language use arXiv preprint. https://arxiv.org/abs/2303.08014.
[24] Liao, W., Liu, Z., Dai, H., Xu, S., Wu, Z., Zhang, Y., Li, X. (2023). Differentiating ChatGPT generated and human written medical texts: Quantitative study. JMIR Medical Education, 9, e48904.
[25] Doru, B., Maier, C., Busse, J. S., Lücke, T., Schönhoff, J., Enax Krumova, E., Hessler, S., Berger, M., Tokic, M. (2025). Detecting artificial intelligence generated versus human written medical student essays: Semirandomized controlled study. JMIR Medical Education, 11, e62779.
[26] Nowacki, L., Wrochna, A. E. (2025). ChatGPT theses. Identifying distinctive markers in AI-generated versus human created texts: A multimodal analysis in university education. E-Learning and Digital Media, 20427530251331083.
[27] Georgiou, G. (2025). Key features to distinguish between human and AI generated texts: What university professors say [Preprint]. https://osf.io/preprints/psyarxiv/jvytm_v2.
[28] Amirjalili, F., Neysani, M., Nikbakht, A. (2024). Exploring the boundaries of authorship: A comparative analysis of AI-generated text and human academic writing in English literature. Frontiers in Education, 9, 1347421. https://doi.org/10.3389/feduc.2024.1347421.
[29] Opara, C. (2025). Distinguishing AI-generated and human written text through psycholinguistic analysis. In A. I. Cristea, E. Walker, Y. Lu, O. C. Santos, S. Isotani (Eds.), Artificial Intelligence in Education (AIED 2025, Lecture Notes in Computer Science, Vol. 15881). Springer. https://doi.org/10.1007/978-3-031-98462-4_27.
[30] Etaat, F. (2026). Exploring linguistic fingerprints in human and AI-generated texts: An NLP-based approach in second language writing. Ampersand, 16, 100258.
[31] Abdulhamed, A. A., Ranjan Singh, P., Xiong, S. (2026). A novel approach for distinguishing human and AI-generated texts. ACM Transactions on Asian and Low-Resource Language Information Processing, 25(7), 1-34.
[32] Fröhling, L., Zubiaga, A. (2021). Feature-based detection of automated language models: Tackling GPT-2, GPT-3 and Grover. PeerJ Computer Science, 7, e443. https://doi.org/10.7717/peerj-cs.443. [33] Liang, W., Yuksekgonul, M., Mao, Y., Wu, E., Zou, J. (2023). GPT detectors are biased against non-native English writers. arXiv preprint. http://arxiv.org/abs/2304.02819.
[34] Adamson, D. (2023, October 26). New research: Turnitin’s AI detector shows no statistically significant bias against English Language Learners. Turnitin. https://www.turnitin.com/blog/new-research-turnitins- ai-detector-shows-no-statistically-significant-bias-against-english-language-learners.
[35] Tian, E. (2023, October 25). ESL Bias in AI Detection is an Outdated Narrative. GPTZero. https:// gptzero.me/news/esl-and-ai-detection.
[36] Elkhatat, A. M., Elsaid, K., Almeer, S. (2023). Evaluating the efficacy of AI content detection tools in differentiating between human and AI-generated text. International Journal for Educational Integrity, 19(1), 17. https://doi.org/10.1007/s40979-023-00140-5.
[37] OpenAI. (2023, January 31). New AI classifier for indicating AI-written text. https://openai.com/blog/ new-ai-classifier-for-indicating-ai-written-text.
[38] Chaka, C. (2023). Detecting AI content in responses generated by ChatGPT, YouChat, and Chatsonic: The case of five AI content detection tools. Journal of Applied Learning and Teaching, 6(2). https://doi.org/ 10.37074/jalt.2023.6.2.12
[39] Gao, C. A., Howard, F. M., Markov, N. S., Dyer, E. C., Ramesh, S., Luo, Y., Pearson, A. T. (2022). Comparing scientific abstracts generated by ChatGPT to original abstracts using an artificial intelligence output detector, plagiarism detector, and blinded human reviewers. bioRxiv. https://doi.org/10.1101/2022.12.23.521610.
[40] Krishna, K., Song, Y., Karpinska, M., Wieting, J., Iyyer, M. (2023). Paraphrasing evades detectors of AIgenerated text, but retrieval is an effective defense. arXiv preprint. http://arxiv.org/abs/2303.13408.
[41] Orenstrakh, M. S., Karnalim, O., Suarez, C. A., Liut, M. (2023). Detecting LLM-generated text in computing education: A comparative study for ChatGPT cases. arXiv preprint. http://arxiv.org/abs/2307.07411.
[42] Perkins, M. (2023). Academic integrity considerations of AI large language models in the post pandemic era: ChatGPT and beyond. Journal of University Teaching Learning Practice, 20(2). https://doi.org/10.53761/ 1.20.02.07.
[43] Walters, W. H. (2023). The effectiveness of software designed to detect AI-generated writing: A comparison of 16 AI text detectors. Open Information Science, 7(1). https://doi.org/10.1515/opis-2022-0158.
[44] Weber-Wulff, D., Anohina-Naumeca, A., Bjelobaba, S., Foltýnek, T., Guerrero-Dib, J., Popoola, O., Šigut, P., Waddington, L. (2023). Testing of detection tools for AI-generated text. International Journal for Educational Integrity, 19(1). https://doi.org/10.1007/s40979-023-00146.
[45] Mitchell, E., Lee, Y., Khazatsky, A., Manning, C. D., Finn, C. (2023). DetectGPT: Zero shot machine generated text detection using probability curvature. arXiv preprint. http://arxiv.org/abs/2301.11305.
[46] Originality.AI. (2023). AI Content Detector Accuracy Review + Open Source Dataset and Research Tool. https://originality.ai
[47] Perkins, M., Roe, J., Postma, D., McGaughran, J., Hickerson, D. (2023). Detection of GPT-4 generated text in higher education: Combining academic judgement and software to identify generative AI tool misuse. Journal of Academic Ethics. https://doi.org/10.1007/s10805-023-09492-6.
[48] Dhaini, M., Poelman, W., Erdogan, E. (2023). Detecting ChatGPT: A survey of the state of detecting ChatGPTgenerated text. In M. Hardalov, B. V. Kancheva, I. Nikolova Koleva, M. Slavcheva (Eds.), Proceedings of the 8th Student Research Workshop associated with the International Conference Recent Advances in Natural Language Processing (p. 1–12). https://aclanthology.org/2023.ranlp-stud.1.
[49] Wu, J. (2025). A corpus based multidimensional analysis of linguistic features between human-authored and ChatGPT-generated compositions. International Journal of Linguistics, Literature and Translation, 8(5), 102–110. https://doi.org/10.32996/ijllt.2025.8.5.10.
[50] Chang, T. A., Bergen, B. K. (2024). Language model behavior: A comprehensive survey. Computational Linguistics, 50(1), 293–350. https://doi.org/10.1162/coli_a_00492.
[51] Sadasivan, V. S., Kumar, A., Balasubramanian, S., Wang, W., Feizi, S. (2025). Can AI-generated text be reliably detected Stress testing AI text detectors under various attacks. Transactions on Machine Learning Research, 1.
[52] Opara, C. (2025). Distinguishing AI-generated and human written text through psycholinguistic analysis. In A. I. Cristea, E. Walker, Y. Lu, O. C. Santos, S. Isotani (Eds.), Artificial Intelligence in Education (AIED 2025, Lecture Notes in Computer Science, Vol. 15881). Springer. https://doi.org/10.1007/978-3-031-98462-4_27 (Note: This is a duplicate entry of [29] but with a different volume number. It has been retained here as per the original numbering).
[53] Georgiou, G. P. (2025). Differentiating between human written and AI-generated texts using automatically extracted linguistic features. Information, 16(11), 979. https://doi.org/10.3390/info16110979.
[54] Kancheva, B. V., Nikolova Koleva, I., Slavcheva, M. (Eds.) (2023). Proceedings of the 8th Student Research Workshop associated with the International Conference Recent Advances in Natural Language Processing. https://aclanthology.org/2023.ranlp-stud.1.