{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T09:32:21Z","timestamp":1777714341832,"version":"3.51.4"},"reference-count":52,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T00:00:00Z","timestamp":1777420800000},"content-version":"vor","delay-in-days":118,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100021856","name":"Ministero dell'Universit\u00e0 e della Ricerca","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100021856","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["International Journal of Intelligent Systems"],"published-print":{"date-parts":[[2026,1]]},"abstract":"<jats:p>The emergence of large language models (LLMs) has significantly advanced natural language processing (NLP); however, their capacity to generate human\u2010like content introduces serious security concerns. In particular, the misuse of LLMs for disinformation and impersonation on social media platforms such as X creates new opportunities for large\u2010scale manipulation and deception of users. This study aims to conduct a comprehensive investigation to (i) understand the distinct stylistic features effectively mimicked by 10 different LLMs and (ii) distinguish between LLM\u2010driven and human authors when LLMs are explicitly instructed to mimic a specific human writing style. In particular, we design adversarial prompts to mimic the writing style of human authors based on key stylometric features (quantitative analysis of writing style) and assess the mimicking effectiveness of different LLMs through extensive statistical testing. In addition, we conduct a survey that gauges human ability to recognize the author of a text and train machine learning models to identify human\u2010 and LLM\u2010driven authors, focusing on scenarios where specifically crafted adversarial prompts are employed to facilitate style impersonation. Our findings demonstrate that, when explicitly instructed, LLMs can effectively replicate features of human writing style. In addition, the survey results indicate that it is challenging for the participants to distinguish between the different types of authors. In fact, the participants only demonstrated a classification accuracy of 15% in correctly identifying the text generated by the LLM. In contrast, high detection performance is achieved only when the training data incorporate adversarially generated LLM samples produced using impersonation\u2010oriented prompts. Under this threat\u2010model\u2013aligned training regime, stylometric\u2010based classifiers exhibit strong discriminative capability, attaining classification accuracy of up to 99% in distinguishing human\u2010authored text from LLM\u2010generated authorship.<\/jats:p>","DOI":"10.1155\/int\/8117975","type":"journal-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T10:10:38Z","timestamp":1777457438000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Breaking the Imitation Game: Can LLMs Fool Humans and Machines Alike?"],"prefix":"10.1155","volume":"2026","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0710-2531","authenticated-orcid":false,"given":"Ubaid","family":"Ullah","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-1256-7076","authenticated-orcid":false,"given":"Sonia","family":"Laudanna","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andrea","family":"Di Sorbo","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Corrado Aaron","family":"Visaggio","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2026,4,29]]},"reference":[{"key":"e_1_2_13_1_2","article-title":"Evaluation of Domain-specific Prompt Engineering Attacks on Large Language Models","volume":"362","author":"Ashcroft C.","year":"2024","journal-title":"ESS Open Archive eprints"},{"key":"e_1_2_13_2_2","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.adh1850"},{"key":"e_1_2_13_3_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2024.09.653"},{"key":"e_1_2_13_4_2","article-title":"Navigating the Web of Disinformation and Misinformation: Large Language Models as Double-Edged Swords","author":"Shah S. B.","year":"2024","journal-title":"IEEE Access"},{"key":"e_1_2_13_5_2","doi-asserted-by":"crossref","unstructured":"MajumdarD. BoyinaP. RayidiS. S. P. SaiY. R. andGangashettyS. V. Beyond Text: Nefarious Actors Harnessing Llms for Strategic Advantage 2024 1\u20137.","DOI":"10.1109\/ISCS61804.2024.10581181"},{"key":"e_1_2_13_6_2","article-title":"Silver Lining in the Fake News Cloud: Can Large Language Models Help Detect Misinformation?","author":"Kumar R.","year":"2024","journal-title":"IEEE Transactions on Artificial Intelligence"},{"key":"e_1_2_13_7_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.mlwa.2024.100545"},{"key":"e_1_2_13_8_2","first-page":"1","volume-title":"IEEE","author":"Cardaioli M.","year":"2021"},{"key":"e_1_2_13_9_2","volume-title":"The Global Risks Report","author":"Wef","year":"2024"},{"key":"e_1_2_13_10_2","doi-asserted-by":"publisher","DOI":"10.20532\/cit.2024.1005778"},{"key":"e_1_2_13_11_2","doi-asserted-by":"crossref","unstructured":"SakibM. N. IslamM. A. PathakR. andArifinM. M. Risks Causes and Mitigations of Widespread Deployments of Large Language Models (LLMs): a Survey 2024 1\u20137.","DOI":"10.1109\/AIBThings63359.2024.10863356"},{"key":"e_1_2_13_12_2","article-title":"Security and Privacy Challenges of Large Language Models: a Survey","author":"Das B. C.","year":"2024","journal-title":"ACM Computing Surveys"},{"key":"e_1_2_13_13_2","doi-asserted-by":"publisher","DOI":"10.1186\/s42400-025-00361-w"},{"key":"e_1_2_13_14_2","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-02157-2","volume-title":"Natural Language Processing for Social Media","author":"Farzindar A.","year":"2015"},{"key":"e_1_2_13_15_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.mlwa.2024.100545"},{"key":"e_1_2_13_16_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2023.106815"},{"key":"e_1_2_13_17_2","volume-title":"How do Humans Perceive Adversarial Text? A Reality Check on the Validity and Naturalness of Word-based Adversarial Attacks","author":"Salijona Dyrmishi M. C.","year":"2023"},{"key":"e_1_2_13_18_2","first-page":"208","volume-title":"Beyond Words: Stylometric Analysis for Detecting AI Manipulation on Social Media","author":"Ullah U.","year":"2024"},{"key":"e_1_2_13_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/3132039"},{"key":"e_1_2_13_20_2","article-title":"How Close is Chatgpt to Human Experts? Comparison Corpus, Evaluation, and Detection","author":"Guo B.","year":"2023","journal-title":"arXiv preprint arXiv:2301.07597"},{"key":"e_1_2_13_21_2","doi-asserted-by":"crossref","unstructured":"ShenX. ChenZ. BackesM. ShenY. andZhangY. \u201d Do Anything Now\u201d: Characterizing and Evaluating in-the-wild Jailbreak Prompts on Large Language Models 2024 1671\u20131685.","DOI":"10.1145\/3658644.3670388"},{"key":"e_1_2_13_22_2","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.2374"},{"key":"e_1_2_13_23_2","doi-asserted-by":"crossref","unstructured":"MoY. WangY. WeiZ. andWangY. Fight Back Against Jailbreaking via Prompt Adversarial Tuning 2024.","DOI":"10.52202\/079017-2049"},{"key":"e_1_2_13_24_2","first-page":"57","volume-title":"LAMPS \u201924","author":"Zhu K.","year":"2024"},{"key":"e_1_2_13_25_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11704-023-2639-2"},{"key":"e_1_2_13_26_2","article-title":"Your Language Model Can Secretly Write like Humans: Contrastive Paraphrase Attacks on LLM-Generated Text Detectors","author":"Fang H.","year":"2025","journal-title":"arXiv preprint arXiv:2505.15337"},{"key":"e_1_2_13_27_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbvi.2023.e00388"},{"key":"e_1_2_13_28_2","doi-asserted-by":"crossref","unstructured":"BhandarkarA. WilsonR. SwarupA. andWoodardD. Emulating Author Style: a Feasibility Study of Prompt-enabled Text Stylization with Off-the-Shelf LLMs 2024 76\u201382.","DOI":"10.18653\/v1\/2024.personalize-1.6"},{"key":"e_1_2_13_29_2","unstructured":"HerboldS. TrautschA. KiktevaZ. andHautli-JaniszA. Large Language Models Can Impersonate Politicians and Other Public Figures 2024."},{"key":"e_1_2_13_30_2","article-title":"Using Prompts to Guide Large Language Models in Imitating a Real Person\u2019s Language Style","author":"Chen Z.","year":"2024","journal-title":"arXiv preprint arXiv:2410.03848"},{"key":"e_1_2_13_31_2","doi-asserted-by":"crossref","unstructured":"WuJ. GuoJ. andHooiB. Fake News in Sheep\u2019s Clothing: Robust Fake News Detection Against LLM-Empowered Style Attacks 2024 3367\u20133378.","DOI":"10.1145\/3637528.3671977"},{"key":"e_1_2_13_32_2","unstructured":"BhatM. M. Study of Effectiveness of Stylometry in Misinformation Detection 2022 The Ohio State University Master\u2019s thesis."},{"key":"e_1_2_13_33_2","unstructured":"PrzystalskiK. Argasi\u0144skiJ. Grabska-Gradzi\u0144skaI. andOchabJ. Stylometry Recognizes Human and Llm-Generated Texts in Short Samples."},{"key":"e_1_2_13_34_2","volume-title":"StyloAI: Distinguishing AI-generated Content with Stylometric Analysis","author":"Opara C.","year":"2024"},{"key":"e_1_2_13_35_2","doi-asserted-by":"crossref","unstructured":"LeeD. XieS. RahmanS. PatK. LeeD. ChenQ. A. andPrompterS.\u201d A Linguistic Approach to Understanding and Detecting Jailbreak Attacks Against Large-Language Models 2023 77\u201387.","DOI":"10.1145\/3689217.3690618"},{"key":"e_1_2_13_36_2","first-page":"1467","volume-title":"Reframing and Broadening Adversarial Stylometry for Academic Integrity","author":"Carmichael J. J.","year":"2024"},{"key":"e_1_2_13_37_2","first-page":"461","volume-title":"IEEE","author":"Afroz S.","year":"2012"},{"key":"e_1_2_13_38_2","doi-asserted-by":"crossref","DOI":"10.1051\/e3sconf\/202337103007","volume-title":"Application of Artificial Intelligence Systems for Stylometric Analysis of Texts as Factor of Sustainable Development","author":"Mikherskii R.","year":"2023"},{"key":"e_1_2_13_39_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i22.34525"},{"key":"e_1_2_13_40_2","article-title":"Fast-Detectgpt: Efficient zero-shot Detection of machine-generated Text via Conditional Probability Curvature","author":"Bao G.","year":"2023","journal-title":"arXiv preprint arXiv:2310.05130"},{"key":"e_1_2_13_41_2","doi-asserted-by":"publisher","DOI":"10.1145\/3715073.3715076"},{"key":"e_1_2_13_42_2","unstructured":"SunM. YinY. XuZ. KolterJ. Z. andLiuZ. Idiosyncrasies in Large Language Models 2025."},{"key":"e_1_2_13_43_2","doi-asserted-by":"crossref","unstructured":"MaiZ. ZhangJ. XuZ. andXiaoZ. Is Llama 3 Good at Sarcasm Detection? a Comprehensive Study 2024 141\u2013145.","DOI":"10.1145\/3696271.3696294"},{"key":"e_1_2_13_44_2","doi-asserted-by":"crossref","unstructured":"CresciS. Di PietroR. PetrocchiM. SpognardiA. andTesconiM. The paradigm-shift of Social Spambots: Evidence Theories and Tools for the Arms Race 2017 963\u2013972.","DOI":"10.1145\/3041021.3055135"},{"key":"e_1_2_13_45_2","doi-asserted-by":"crossref","DOI":"10.1002\/9780470479216.corpsy0524","article-title":"Mann-Whitney U Test","author":"McKnight P. E.","year":"2010","journal-title":"The Corsini Encyclopedia of Psychology"},{"key":"e_1_2_13_46_2","volume-title":"Effect Sizes for Research: A Broad Practical Approach","author":"Grissom R. J.","year":"2005"},{"key":"e_1_2_13_47_2","doi-asserted-by":"publisher","DOI":"10.1145\/571681.571686"},{"key":"e_1_2_13_48_2","volume-title":"Surveying the Social World: Principles and Practice in Survey Research","author":"Aldridge A.","year":"2001"},{"key":"e_1_2_13_49_2","volume-title":"Questionnaire Design, Interviewing and Attitude Measurement","author":"Oppenheim A. N.","year":"2000"},{"key":"e_1_2_13_50_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2021.115222"},{"key":"e_1_2_13_51_2","doi-asserted-by":"publisher","DOI":"10.1155\/2019\/9431073"},{"key":"e_1_2_13_52_2","doi-asserted-by":"crossref","unstructured":"GilaniZ. KochmarE. andCrowcroftJ. Classification of Twitter Accounts into Automated Agents and Human Users 2017 489\u2013496.","DOI":"10.1145\/3110025.3110091"}],"container-title":["International Journal of Intelligent Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1155\/int\/8117975","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1155\/int\/8117975","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1155\/int\/8117975","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T10:10:56Z","timestamp":1777457456000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1155\/int\/8117975"}},"subtitle":[],"editor":[{"given":"Richard","family":"Murray","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,1]]},"references-count":52,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1]]}},"alternative-id":["10.1155\/int\/8117975"],"URL":"https:\/\/doi.org\/10.1155\/int\/8117975","archive":["Portico"],"relation":{},"ISSN":["0884-8173","1098-111X"],"issn-type":[{"value":"0884-8173","type":"print"},{"value":"1098-111X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1]]},"assertion":[{"value":"2025-10-25","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-27","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-29","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"8117975"}}