V.I. Budzko1, V.G. Belenkov2, P.A. Keyer3

1–3 Federal Research Center «Computer Science» of the Russian Academy Sciences (Moscow, Russia)
1 vbudzko@ipiran.ru, 2 vbelenkov@ipiran.ru, 3 pkeyer@ipiran.ru

In the context of the ever-increasing role of automated information support tools in various fields of activity, disruption or termination of such support can lead to serious disruptions to the corresponding business processes. Automated data intensive domains systems (DID systems) mostly belong to the class of critical information infrastructure systems. The implementation of cyber threats can lead to catastrophic consequences. Therefore, ensuring a high level of cybersecurity is one of the priority tasks, for the solution of which artificial intelligence (AI) and machine learning (ML) methods are increasingly used. The article discusses new solutions for automating the processes of ensuring cybersecurity of DID systems using AI and ML. The development of cybersecurity of DID systems is increasingly associated with the massive use of generative neural networks (GNN). The paper lists the main tasks of ensuring cybersecurity, where GNN are used, technologies for their use. The advantage of using the development of GNN – generative adversarial networks (GAN) in the preparation of training data sets is shown. The solutions ensuring the success of their application, the areas of application and the tasks for which it is effective to use the GAN, and the prospects for their further use are defined. The issues of ensuring the necessary qualification of personnel associated with the functioning of the cybersecurity system, the issues of its own security throughout its life cycle are considered.

1. Budzko V.I., Korolev V.I., Belenkov V.G., Kejer P.A. Kiberbezopasnost` sistem, realizuyushhix intensivnoe ispol`zovanie danny`x. Mesto kiberbezopasnosti v zashhite in-formacii. Sistemy` vy`sokoj dostupnosti. 2024. T. 20. № 1. S. 16–29. DOI: https://doi.org/10.18127/ j20729472-202401-02
2. Budzko V.I., Medennikov V.I., Korolev V.I., Belenkov V.G., Kejer P.A. Osobennosti sistem s intensivny`m ispol`zovaniem danny`x kak ob``ektov obespecheniya bezopasnosti informacii na primere ASU agrotexnologicheskimi processami. Sistemy` vy`sokoj dostupnosti. 2024. T. 20. № 2. S. 28–39. DOI: https://doi.org/10.18127/j20729472-202402-02
3. Budzko V.I., Korolev V.I., Belenkov V.G., Kejer P.A. Kiberbezopasnost` sistem, realizuyushhix intensivnoe ispol`zovanie danny`x.
   Bezopasnost` informacionnoj infra-struktury` oblasti s intensivny`m ispol`zovaniem danny`x. Sistemy` vy`sokoj dostupnosti. 2024. T. 20. № 3. S. 5–18. DOI: https://doi.org/10.18127/j20729472-202403-01
4. Budzko V.I., Korolev V.I., Belenkov V.G., Kejer P.A. Kiberbezopasnost` sistem, realizuyushhix intensivnoe ispol`zovanie danny`x.
   Bezopasnost` Raspredelennoj Sredy` sistemy`, realizuyushhej intensivnoe ispol`zovanie danny`x. Sistemy` vy`sokoj dostupnosti. 2024. T. 20. № 4. S. 15–23. DOI: https://doi.org/10.18127/j20729472-202404-02
5. Budzko V.I., Belenkov V.G. Kiberbezopasnost` sistem, realizuyushhix intensivnoe ispol`zovanie danny`x, ispol`zuyushhix texnologii iskusstvennogo intellekta. Sistemy` vy`sokoj dostupnosti. 2025. T. 21. № 1. S. 52–62. DOI: https://doi.org/10.18127/j20729472-202501-05
6. Proshina M. Kursy` i materialy` ot SF Education. Kak rabotaet nejroset` https://blog.sf.education/kak-rabotaet-neiroset/
7. Shavlyugo N., Shpringer L., Ovchinnikova P. Analiz danny`x 11 iyulya 2023 https://practicum.yandex.ru/blog/mashinnoe-obuchenie-s-uchitelem-i-bez/
8. Maly`shev I.O., Smirnov A.A. Obzor sovremenny`x generativny`x nejrosetej: otechestvennaya i zarubezhnaya praktika. International Journal of Humanities and Natural Sciences. 2024. V. 1-2 (88). P. 168–171. DOI: 10.24412/2500-1000-2024-1-2-168-171.
9. Sidoruk D., Vixreva M. Chto takoe generativny`j iskusstvenny`j intellekt i kak on rabotaet. https://practicum.yandex.ru/blog/ generativnyi-iskusstvennyi-intellekt/
10. Goodfellow I. and al. Generative adversarial nets. In Proc. Adv. Neural Inf. Process. Syst. 2014. V. 27. P. 1–14.
11. Wang K. and al. Generative adversarial networks: Introduction and outlook. IEEE/CAA J. Autom. Sinica. 2017. V. 4. № 4. P. 588–598.
12. Arzamascev N.A. Osobennosti ispol`zovaniya iskusstvenny`x nejronny`x setej v sfere informacionnoj bezopasnosti. Nauchno-obrazovatel`ny`j zhurnal dlya studentov i prepodavatelej «StudNet». 2022. № 5. S. 3936–3945.
13. Chto takoe kiber-killer-cepochka i kak ona rabotaet {e`tapy` i primery`}. https://ru.info-sec.wiki/?p=277
14. Celevy`e ataki: e`tapy`, instrumenty`, metody`. https://www.sberbank.ru/ru/person/kibrary/articles/celevye-ataki-ehtapy-instrumenty-metody
15. Iskusstvenny`j intellekt i mashinnoe obuchenie v kiberbezopasnosti – prognoz na budushhee. https://www.kaspersky.ru/resource-center/definitions/ai-cybersecurity
16. https://neurohive.io/ru/osnovy-data-science/gan-rukovodstvo-dlja-novichkov/
17. Kiberbezopasnost` i iskusstvenny`j intellekt: v chem sila, brat? https://habr.com/ru/companies/cdnnow/articles/816185/
18. Istoriya goda: vliyanie iskusstvennogo intellekta na kiberbezopasnost`. https://securelist.ru/story-of-the-year-2023-ai-impact-on-cybersecurity/108558/
19. Nguyen X.-H. and al. Preventing Adversarial Attacks Against Deep Learning-Based Intrusion Detection System. Proc. of 17th International Conference Information Security Practice and Experience. ISPEC 2022. Taipei. Taiwan. November 23–25. 2022. P. 382–396.
20. Nath B. and al. Data Mining for the Security of Cyber Physical Systems Using Deep Learning Methods. Proc. of the 17th International Conference on Cyber Warfare and Security. Albany. New York. USA. 17–18 March 2022. P. 591–598.
21. Crouse M. Generative AI Defined: How It Works, Benefits, and Limitations. Published October 24. 2024. https://www.techrepublic.com/ article/what-is-generative-ai/
22. Generativny`j II i kiberbezopasnost`: preimushhestva i problemy`. https://www.itweek.ru/ai/article/detail.php?ID=226224
23. Dunmore A. and al. A Comprehensive Survey of Generative Adversarial Networks (GANs) in Cybersecurity Intrusion Detection. IEEE Access VOLUME 11. 2023. P. 76071-76094. Digital Object Identifier 10.1109/ACCESS.2023.3296707 https://www.researchgate.net/publication/372404301_A_Comprehensive_Survey_of_Generative_Adversarial_Networks_GANs_in_Cybersecurity_Intrusion_Detection
24. Chen Li and al. MAD-GAN: multivariate anomaly detection for time series data with genera-tive adversarial networks. International Conference on Artificial Neural Networks. 2019. P. 703–716. https://doi.org/10.1007/978-3-030-30490-456
25. Quyen N.H. and al. Federated Intrusion Detection on Non-IID Data for IIV Networks Using Generative Adversarial Networks and Reinforcement Learning. Proc. of 17th International Conference Information Security Practice and Experience. ISPEC 2022. Taipei. Taiwan. November 23–25. 2022. P. 364–381.
26. GOST R 59895-2021. Texnologii iskusstvennogo intellekta v obrazovanii. Obshhie polozheniya i terminologiya.