Indicators of Texts Generated by Neural Networks in the Aspect of Authorship Examination (Based on Experimental Study)

   Diagnostics of indicators of texts generated by neural networks in the aspect of forensic authorship is a relevant and complex task, the solution of which requires new skills from the expert. The article provides an overview of the recent publications in the field of identifying the indicators of generation of Russian-language texts at the current stage of development of science and technology. The article presents the results of a study conducted at the Forensic Center of the Ministry of Internal Affairs of Russia as part of an All-Russian training workshop among experts in the field of forensic authorship examination of forensic units of the Ministry of Internal Affairs of Russia. For the first time, using the material of colloquial texts within the law enforcement discourse, generated using freely available programs (GPT4, DeepSeek, Grok 2, Hailuo), through expert analysis, the signs indicating the use of these programs are summarized. They were formed on the basis of their reproducibility in the conclusions obtained by forensic authorship experts with a subsequent assessment of their relevance. The identified indicators expand the list proposed in existing scientific works. Attention is also paid to the relationship between the conditions for creating texts set by programs and the quality of implementation of each of the specified parameters in the generated text. The article outlines the feasibility of developing an interdisciplinary approach to expert analysis of such texts, including methods of computer and classical linguistics. An “expert experiment” traditionally used in the production of certain types of forensic examinations is proposed as a new method for authorship examination. In the context of intensive development of technologies for creating human-quality texts, the need for dynamic improvement of the scientific and methodological base and further research is outlined.

1. Litvinova T.A., Mikros J., Dekhnich O.V. Writing in the Era of Large Language Models: A Bibliometric Analysis of Research Field. Research Result. Theoretical and Applied Linguistics. 2024. Vol. 10. No. 4. P. 5–16. (In Russ.). doi: 10.18413/2313-8912-2024-10-4-0-1

2. Telpov R.E., Lartsina S.V. Typological Differences of Natural and Neural Network-Generated Texts in a Quantitative Aspect. Nauchnyi Dialog. 2023. No. 12 (7). P. 47–65. (In Russ.). doi: 10.24224/2227-1295-2023-12-7-47-65

3. Aidagulova A.R. Features of Texts Generated by Artificial Intelligence. Bulletin of the Bashkir State Pedagogical University named after M. Akmulla. 2023. No. 4 (72). P. 154–156. (In Russ.).

4. Cherkasova M.N., Taktarova A.V. Attributes of Generated Text in Academic Discourse: The Problem of Identification. Philology. Theory & Practice. 2024. Vol. 17. Issue 7. (In Russ.). doi: 10.30853/phil20240307

5. Olomskaya N.N., Yurova E.A. Linguopragmatic Features of AI-Generated Text in the Media Discourse of Social Networks (on the Example of Texts Devoted to the Governor Election in the Nizhny Novgorod Region, Russian Federation, 2023). Russian Social and Humanitarian Journal. 2025. No. 2. (In Russ.). doi: 10.18384/2224-0209-2025-2-1649

6. Kolmogorova A.V., Margolina A.V. Written vs. Generated Text: “Naturalness” as a Textual and Psycholinguistic Category. Research Result. Theoretical and Applied Linguistics. 2024. Vol. 10. No. 2. P. 71–99. (In Russ.). doi: 10.18413/2313-8912-2024-10-2-0-4

7. Osetrova E.V., Sedova A.V. Characteristics of the Generated Text: Linguistic and Socio-Communicative Analysis. Siberian Philological Forum. 2025. No. 2 (31). P. 45–55. (In Russ.).

8. Gromova A.V., Loginova S.N., Kitaeva E.S., Oshkukov S.S., Manyanin P.A. On Identifying Signs of Artificial Text Generation at the Stage of Determining the Suitability of Objects for Conducting Authorship Research: Information Letter. Moscow: EKTs MVD Rossii, 2020. 37 p. (In Russ.).

9. Gromova A.V., Loginova S.N., Kitaeva E.S. On the Issue of Identifying Signs of Artificial Text Generation When Conducting Authorship Research. Fundamental Linguistics and Problems of Forensic Examination: Social Networks as an Object of Scientific and Expert Analysis : Collection of Scientific Papers Following the Results of the International Scientific Conference (Moscow, October 5–6, 2021). Moscow: Gosudarstvennyi institut russkogo yazyka im. A.S. Pushkina, 2022. P. 11–16. (In Russ.).

10. Gromova A.V. Correspondence in the Messenger: Identifying the Author by Text in the Context of the Transformation of Individualizing Features. Science Journal of Volgograd State University. Linguistics. 2021. Vol. 20. No. 2. P. 87–98. (In Russ.). doi: 10.15688/jvolsu2.2021.2.8

11. Gromova A.V., Manyanin P.A., Rostovskaya A.V., Oshkukov S.S., Berdnikova M.A., Khomyakov Yu.A., Tayurskaya M.Yu. Author Identification by Texts Created during Correspondence in a Messenger. Methodological Recommendations. Moscow: EKTs MVD Russii, 2022. 108 p. (In Russ.).

12. Kazmin V.V., Gromova A.V. How to Recognize Artificial Speech. Police of Russia. 2024. No. 4. P. 12–14. (In Russ.).

13. Gromova A.V., Berdnikova M.A. Signs of Dialogical Printed Speech Generated by Chatbots: An Experimental Study. Expert-Criminalist. 2025. No. 3. P. 5–9. (In Russ.).

14. Kazmin V.V. New Approaches to Improving the Qualifications of Specialists Engaged in the Production of Examinations to Combat Cybercrime. Professional: Practical Almanac of the Ministry of Internal Affairs of Russia. 2025. No. 1. P. 20–21. (In Russ.).

15. Litvinova T.A., Gromova A.V. Current Problems of Forensic Authorship Analysis and the Possibility of Their Solution with the Use of Computer Methods: Problems and Prospects. Science Journal of Volgograd State University. Linguistics. 2020. Vol. 19. No. 1. P. 77–88. (In Russ.).

16. Oshkukov S.S. Stylometric Identification of the Author of the Text in Authorship Examination. Bulletin of the Voronezh Institute of the Ministry of Internal Affairs of Russia. 2023. No. 1. P. 89–97. (In Russ.).