НЕКОТОРЫЕ АСПЕКТЫ ПРИМЕНЕНИЯ ТЕХНОЛОГИЙ ИСКУССТВЕННОГО ИНТЕЛЛЕКТА В ЗАДАЧАХ ЗАЩИТЫ ИНФОРМАЦИИ (ОБЗОР)
Ключевые слова:
Информационная безопасность, кибербезопасность, технический канал утечки, искусственный интеллект, доверенный искусственный интеллект
Аннотация
Технологии искусственного интеллекта (ИИ) являются одной из наиболее динамично разви-
вающихся областей обработки информации. Технологии ИИ используются как для обеспечения
защиты информации, так и для организации атак на средства ее защиты. Сами системы ИИ
могут содержать уязвимости и быть подвержены атакам различного рода. В статье анализи-
руются некоторые аспекты применения технологий ИИ в задачах защиты информации. В рамках
задачи биометрической идентификации рассматриваются угрозы подделки биометрических
идентификационных признаков с целью получения прав доступа, и способы противодействия та-
ким угрозам. Анализируются преимущества использования ИИ при защите информации в компь-
ютерных системах и сетях по сравнению с традиционными средствами защиты. На примере аку-
стического канала утечки информации от клавиатуры иллюстрируется использование техноло-
гий ИИ для обработки данных из технических каналов утечки. Рассматриваются методы повы-
шения информативности таких каналов, использующие временные сверточные сети и модели
классификации изображений, а также способы противодействия им. Отдельное внимание уделе-
но вопросам информационной безопасности в набирающих популярность системах сжатия и пе-
редачи информации без значительных смысловых потерь (направление Semantic Communications).
Рассматриваются ряд вопросов информационной безопасности, возникающих при использовании
больших языковых моделей типа ChatGPT, способных массово генерировать уникальный «челове-
коподобный» контент и использовать его для организации фишинговых и других атак социальной
инженерии. Описана атака на системы ИИ с использованием скрытого канала. Уделено внимание
необходимости развития технологий доверенного искусственного интеллекта.
Литература
1. Natsional'naya strategiya razvitiya iskusstvennogo intellekta na period do 2030 goda, utverzhdena
Ukazom Prezidenta Rossiyskoy Federatsii ot 10 oktyabrya 2019 g. № 490 [National Strategy for the
Development of Artificial Intelligence through 2030, approved by the Decree of the President of the
Russian Federation, October 10, 2019, No. 490].
2. McCorduck P., Cfe C. Machines who think: A personal inquiry into the history and prospects of artificial
intelligence. AK Peters/CRC Press, 2004.
3. Mughal A.A. Artificial Intelligence in Information Security: Exploring the Advantages, Challenges,
and Future Directions, Journal of Artificial Intelligence and Machine Learning in Management, 2018,
Vol. 2, No. 1, pp. 22-34.
4. Materialy pervogo foruma «Tsifrovaya ekonomika. Tekhnologii doverennogo iskusstvennogo
intellekta». Moskva, MGU, klaster «Lomonosov», 23 maya 2023 g. [Proceedings of the first forum
"Digital Economy. Technologies of Trusted Artificial Intelligence". Moscow, Moscow State University,
Lomonosov Cluster, May 23, 2023]. Available at: https://ib-bank.ru/trust-ai/materials23 (accessed
on 10 October 2024).
5. Meshcheryakov R.V., Mel'nikov S.Yu., Peresypkin V.A., Khorev A.A. Perspektivnye napravleniya
primeneniya tekhnologiy iskusstvennogo intellekta pri zashchite informatsii [Promising areas of application
of artificial intelligence technologies in information security], Voprosy kiberbezopasnosti
[Cybersecurity Issues], 2024, No. 4 (62), pp. 2-12. DOI: 10.21681/2311-3456-2024-4-02-12. EDN
GJWQWP.
6. Shelke P., Hämäläinen T. Analysing Multidimensional Strategies for Cyber Threat Detection in Security
Monitoring, In M. Lehto, & M. Karjalainen (Eds.), Proceedings of the 23rd European Conference
on Cyber Warfare and Security, 2024, 23, pp. 780-787. Academic Conferences International Ltd. Proceedings
of the European Conference on Cyber Warfare and Security. Available at:
https://doi.org/10.34190/eccws.23.1.2123 (accessed 10 October 2024).
7. Avetisyan A.I. Kiberbezopasnost' v kontekste iskusstvennogo intellekta [Cybersecurity in the Context
of Artificial Intelligence],Vestnik Rossiyskoy akademii nauk [Bulletin of the Russian Academy of Sciences],
2022, Vo.. 92, No. 12, pp. 1119-1123. DOI: 10.31857/S0869587322120039. EDN RYZRRU.
8. Camacho N.G. The Role of AI in Cybersecurity: Addressing Threats in the Digital Age, Journal of
Artificial Intelligence General science (JAIGS), 2024, Vol. 3, No. 1, pp. 143-154. ISSN 3006-4023.
9. Panoff M. et al. A review and comparison of AI-enhanced side channel analysis, ACM Journal on
Emerging Technologies in Computing Systems (JETC), 2022, Vol. 18, No. 3, pp. 1-20.
10. Zhuang L., Zhou F., Tygar J.D. Keyboard acoustic emanations revisited, ACM Transactions on Information
and System Security (TISSEC), 2009, Vol. 13, No. 1, pp. 1-26.
11. Taheritajar A., Harris Z. M., Rahaeimehr R. A Survey on Acoustic Side Channel Attacks on Keyboards,
arXiv preprint arXiv:2309.11012, 2023.
12. Harrison J., Toreini E., Mehrnezhad M. A practical deep learning-based acoustic side channel attack
on keyboards, 2023 IEEE European Symposium on Security and Privacy Workshops (EuroS&PW).
IEEE, 2023, pp. 270-280.
13. Spata M. O. et al. A New Deep Learning Pipeline for Acoustic Attack on Keyboards, IntelliSys 2024.
Cham: Springer Nature Switzerland, 2024, pp. 402-414.
14. Mel'nikov S.Yu., Peresypkin V.A. Ob evolyutsii klassicheskikh veroyatnostnykh modeley yazyka v
estestvenno-yazykovykh prilozheniyakh [On the evolution of classical probabilistic language models
in natural language applications], Vestnik sovremennykh tsifrovykh tekhnologiy [Bulletin of modern
digital technologies], 2023, No. 16, pp. 4-14. EDN YDIGDT.
15. Rodrigues D. et al. A Prototype for Generating Random Key Sounds to Prevent Keyboard Acoustic
Side-Channel Attacks, 2024 IEEE 22nd Mediterranean Electrotechnical Conference (MELECON).
IEEE, 2024, pp. 1287-1292.
16. Yang W. et al. Semantic communications for future internet: Fundamentals, applications, and challenges,
IEEE Communications Surveys & Tutorials, 2022, Vol. 25, No. 1, pp. 213-250.
17. Wang Y. Semantic Communication Networks Empowered Artificial Intelligence of Things, 2024 IEEE
Annual Congress on Artificial Intelligence of Things (AIoT). IEEE, 2024, pp. 189-193.
18. Xie H. et al. Deep learning enabled semantic communication systems, IEEE Transactions on Signal
Processing, 2021, Vol. 69, pp. 2663-2675.
19. Bourtsoulatze E., Kurka D. B., Gündüz D. Deep joint source-channel coding for wireless image transmission,
IEEE Transactions on Cognitive Communications and Networking, 2019, Vol. 5, No. 3,
pp. 567-579.
20. Luo X. et al. Encrypted semantic communication using adversarial training for privacy preserving,
IEEE Communications Letters, 2023, Vol. 27, No. 6, pp. 1486-1490.
21. Nguyen V.L. et al. Security and privacy for 6G: A survey on prospective technologies and challenges,
IEEE Communications Surveys & Tutorials, 2021, Vol. 23, No. 4, pp. 2384-2428.
22. Li Y. et al. Secure Semantic Communications: From Perspective of Physical Layer Security, IEEE
Communications Letters, 2024. DOI: 10.1109/LCOMM.2024.3452715.
23. Hazell J. Large language models can be used to effectively scale spear phishing campaigns, arXiv
preprint arXiv:2305.06972, 2023.
24. Greco F. et al. David versus Goliath: Can Machine Learning Detect LLM-Generated Text? A Case
Study in the Detection of Phishing Emails, ITASEC 2024: The Italian Conference on CyberSecurity,
Italy. CEUR-WS Vol. 3731, 2024.
25. Bylevskiy P.G. Sotsial'no-kul'turnye riski mul'timodal'nykh bol'shikh generativnykh modeley
«iskusstvennogo intellekta» (GenAI) [Socio-cultural risks of multimodal large generative models of
"artificial intelligence" (GenAI)], Kul'tura i iskusstvo [Culture and Art], 2024, No. 6, pp. 213-224.
DOI: 10.7256/2454-0625.2024.6.70926. EDN: DWMERQ.
26. Hanley H.W.A., Durumeric Z. Machine-made media: Monitoring the mobilization of machinegenerated
articles on misinformation and mainstream news websites, Proceedings of the International
AAAI Conference on Web and Social Media, 2024, Vol. 18, pp. 542-556.
27. Simon F.M., Altay S., Mercier H. Misinformation reloaded? Fears about the impact of generative AI on
misinformation are overblown, Harvard Kennedy School Misinformation Review, 2023, Vol. 4, No. 5.
28. Liu Y. et al. ArguGPT: evaluating, understanding and identifying argumentative essays generated by
GPT models, arXiv preprint arXiv:2304.07666, 2023.
29. Wu J. et al. A survey on llm-gernerated text detection: Necessity, methods, and future directions,
arXiv preprint arXiv:2310.14724, 2023.
30. Ghosal S.S. et al. A Survey on the Possibilities & Impossibilities of AI-generated Text Detection,
Transactions on Machine Learning Research, No. 1, 2024.
31. Sadasivan V.S. et al. Can AI-generated text be reliably detected?, arXiv preprint arXiv:2303.11156,
2023.
32. Marshalko G.B., Romanenkov R.A., Trufanova Yu.A. Analiz bezopasnosti proekta natsional'nogo standarta
«Neyrosetevye algoritmy v zashchishchennom ispolnenii. Avtomaticheskoe obuchenie neyrosetevykh
modeley na malykh vyborkakh v zadachakh klassifikatsii» [Security analysis of the draft national standard
"Neural network algorithms in secure implementation. Automatic training of neural network models on
small samples in classification problems"], Tr. Instituta sistemnogo programmirovaniya RAN [Proceedings
of the Institute for System Programming of the Russian Academy of Sciences], 2023, Vol. 35, No. 6,
pp. 179-188. DOI: 10.15514/ISPRAS-2023-35(6)-11. EDN HNDIYD.
33. Grusho A.A. Skrytye kanaly i bezopasnost' informatsii v komp'yuternykh sistemakh [Covert channels
and information security in computer systems], Diskretnaya matematika [Discrete Mathematics],
1998, Vol. 10, No. 1, pp. 3-9.
34. Weiss R. et al. What Was Your Prompt? A Remote Keylogging Attack on AI Assistants, arXiv preprint
arXiv:2403.09751, 2024.
35. Turdakov D.Yu., Avetisyan A.I., Arkhipenko K.V. [i dr.]. Doverennyy Iskusstvennyy intellekt: vyzovy i
perspektivnye resheniya [Trusted Artificial Intelligence: Challenges and Promising Solutions],
Doklady Rossiyskoy akademii nauk. Matematika, informatika, protsessy upravleniya [Reports of the
Russian Academy of Sciences. Mathematics, informatics, control processes], 2022, Vol. 508, No. 1,
pp. 13-18. DOI: 10.31857/S2686954322070207. EDN CVIVCS
Ukazom Prezidenta Rossiyskoy Federatsii ot 10 oktyabrya 2019 g. № 490 [National Strategy for the
Development of Artificial Intelligence through 2030, approved by the Decree of the President of the
Russian Federation, October 10, 2019, No. 490].
2. McCorduck P., Cfe C. Machines who think: A personal inquiry into the history and prospects of artificial
intelligence. AK Peters/CRC Press, 2004.
3. Mughal A.A. Artificial Intelligence in Information Security: Exploring the Advantages, Challenges,
and Future Directions, Journal of Artificial Intelligence and Machine Learning in Management, 2018,
Vol. 2, No. 1, pp. 22-34.
4. Materialy pervogo foruma «Tsifrovaya ekonomika. Tekhnologii doverennogo iskusstvennogo
intellekta». Moskva, MGU, klaster «Lomonosov», 23 maya 2023 g. [Proceedings of the first forum
"Digital Economy. Technologies of Trusted Artificial Intelligence". Moscow, Moscow State University,
Lomonosov Cluster, May 23, 2023]. Available at: https://ib-bank.ru/trust-ai/materials23 (accessed
on 10 October 2024).
5. Meshcheryakov R.V., Mel'nikov S.Yu., Peresypkin V.A., Khorev A.A. Perspektivnye napravleniya
primeneniya tekhnologiy iskusstvennogo intellekta pri zashchite informatsii [Promising areas of application
of artificial intelligence technologies in information security], Voprosy kiberbezopasnosti
[Cybersecurity Issues], 2024, No. 4 (62), pp. 2-12. DOI: 10.21681/2311-3456-2024-4-02-12. EDN
GJWQWP.
6. Shelke P., Hämäläinen T. Analysing Multidimensional Strategies for Cyber Threat Detection in Security
Monitoring, In M. Lehto, & M. Karjalainen (Eds.), Proceedings of the 23rd European Conference
on Cyber Warfare and Security, 2024, 23, pp. 780-787. Academic Conferences International Ltd. Proceedings
of the European Conference on Cyber Warfare and Security. Available at:
https://doi.org/10.34190/eccws.23.1.2123 (accessed 10 October 2024).
7. Avetisyan A.I. Kiberbezopasnost' v kontekste iskusstvennogo intellekta [Cybersecurity in the Context
of Artificial Intelligence],Vestnik Rossiyskoy akademii nauk [Bulletin of the Russian Academy of Sciences],
2022, Vo.. 92, No. 12, pp. 1119-1123. DOI: 10.31857/S0869587322120039. EDN RYZRRU.
8. Camacho N.G. The Role of AI in Cybersecurity: Addressing Threats in the Digital Age, Journal of
Artificial Intelligence General science (JAIGS), 2024, Vol. 3, No. 1, pp. 143-154. ISSN 3006-4023.
9. Panoff M. et al. A review and comparison of AI-enhanced side channel analysis, ACM Journal on
Emerging Technologies in Computing Systems (JETC), 2022, Vol. 18, No. 3, pp. 1-20.
10. Zhuang L., Zhou F., Tygar J.D. Keyboard acoustic emanations revisited, ACM Transactions on Information
and System Security (TISSEC), 2009, Vol. 13, No. 1, pp. 1-26.
11. Taheritajar A., Harris Z. M., Rahaeimehr R. A Survey on Acoustic Side Channel Attacks on Keyboards,
arXiv preprint arXiv:2309.11012, 2023.
12. Harrison J., Toreini E., Mehrnezhad M. A practical deep learning-based acoustic side channel attack
on keyboards, 2023 IEEE European Symposium on Security and Privacy Workshops (EuroS&PW).
IEEE, 2023, pp. 270-280.
13. Spata M. O. et al. A New Deep Learning Pipeline for Acoustic Attack on Keyboards, IntelliSys 2024.
Cham: Springer Nature Switzerland, 2024, pp. 402-414.
14. Mel'nikov S.Yu., Peresypkin V.A. Ob evolyutsii klassicheskikh veroyatnostnykh modeley yazyka v
estestvenno-yazykovykh prilozheniyakh [On the evolution of classical probabilistic language models
in natural language applications], Vestnik sovremennykh tsifrovykh tekhnologiy [Bulletin of modern
digital technologies], 2023, No. 16, pp. 4-14. EDN YDIGDT.
15. Rodrigues D. et al. A Prototype for Generating Random Key Sounds to Prevent Keyboard Acoustic
Side-Channel Attacks, 2024 IEEE 22nd Mediterranean Electrotechnical Conference (MELECON).
IEEE, 2024, pp. 1287-1292.
16. Yang W. et al. Semantic communications for future internet: Fundamentals, applications, and challenges,
IEEE Communications Surveys & Tutorials, 2022, Vol. 25, No. 1, pp. 213-250.
17. Wang Y. Semantic Communication Networks Empowered Artificial Intelligence of Things, 2024 IEEE
Annual Congress on Artificial Intelligence of Things (AIoT). IEEE, 2024, pp. 189-193.
18. Xie H. et al. Deep learning enabled semantic communication systems, IEEE Transactions on Signal
Processing, 2021, Vol. 69, pp. 2663-2675.
19. Bourtsoulatze E., Kurka D. B., Gündüz D. Deep joint source-channel coding for wireless image transmission,
IEEE Transactions on Cognitive Communications and Networking, 2019, Vol. 5, No. 3,
pp. 567-579.
20. Luo X. et al. Encrypted semantic communication using adversarial training for privacy preserving,
IEEE Communications Letters, 2023, Vol. 27, No. 6, pp. 1486-1490.
21. Nguyen V.L. et al. Security and privacy for 6G: A survey on prospective technologies and challenges,
IEEE Communications Surveys & Tutorials, 2021, Vol. 23, No. 4, pp. 2384-2428.
22. Li Y. et al. Secure Semantic Communications: From Perspective of Physical Layer Security, IEEE
Communications Letters, 2024. DOI: 10.1109/LCOMM.2024.3452715.
23. Hazell J. Large language models can be used to effectively scale spear phishing campaigns, arXiv
preprint arXiv:2305.06972, 2023.
24. Greco F. et al. David versus Goliath: Can Machine Learning Detect LLM-Generated Text? A Case
Study in the Detection of Phishing Emails, ITASEC 2024: The Italian Conference on CyberSecurity,
Italy. CEUR-WS Vol. 3731, 2024.
25. Bylevskiy P.G. Sotsial'no-kul'turnye riski mul'timodal'nykh bol'shikh generativnykh modeley
«iskusstvennogo intellekta» (GenAI) [Socio-cultural risks of multimodal large generative models of
"artificial intelligence" (GenAI)], Kul'tura i iskusstvo [Culture and Art], 2024, No. 6, pp. 213-224.
DOI: 10.7256/2454-0625.2024.6.70926. EDN: DWMERQ.
26. Hanley H.W.A., Durumeric Z. Machine-made media: Monitoring the mobilization of machinegenerated
articles on misinformation and mainstream news websites, Proceedings of the International
AAAI Conference on Web and Social Media, 2024, Vol. 18, pp. 542-556.
27. Simon F.M., Altay S., Mercier H. Misinformation reloaded? Fears about the impact of generative AI on
misinformation are overblown, Harvard Kennedy School Misinformation Review, 2023, Vol. 4, No. 5.
28. Liu Y. et al. ArguGPT: evaluating, understanding and identifying argumentative essays generated by
GPT models, arXiv preprint arXiv:2304.07666, 2023.
29. Wu J. et al. A survey on llm-gernerated text detection: Necessity, methods, and future directions,
arXiv preprint arXiv:2310.14724, 2023.
30. Ghosal S.S. et al. A Survey on the Possibilities & Impossibilities of AI-generated Text Detection,
Transactions on Machine Learning Research, No. 1, 2024.
31. Sadasivan V.S. et al. Can AI-generated text be reliably detected?, arXiv preprint arXiv:2303.11156,
2023.
32. Marshalko G.B., Romanenkov R.A., Trufanova Yu.A. Analiz bezopasnosti proekta natsional'nogo standarta
«Neyrosetevye algoritmy v zashchishchennom ispolnenii. Avtomaticheskoe obuchenie neyrosetevykh
modeley na malykh vyborkakh v zadachakh klassifikatsii» [Security analysis of the draft national standard
"Neural network algorithms in secure implementation. Automatic training of neural network models on
small samples in classification problems"], Tr. Instituta sistemnogo programmirovaniya RAN [Proceedings
of the Institute for System Programming of the Russian Academy of Sciences], 2023, Vol. 35, No. 6,
pp. 179-188. DOI: 10.15514/ISPRAS-2023-35(6)-11. EDN HNDIYD.
33. Grusho A.A. Skrytye kanaly i bezopasnost' informatsii v komp'yuternykh sistemakh [Covert channels
and information security in computer systems], Diskretnaya matematika [Discrete Mathematics],
1998, Vol. 10, No. 1, pp. 3-9.
34. Weiss R. et al. What Was Your Prompt? A Remote Keylogging Attack on AI Assistants, arXiv preprint
arXiv:2403.09751, 2024.
35. Turdakov D.Yu., Avetisyan A.I., Arkhipenko K.V. [i dr.]. Doverennyy Iskusstvennyy intellekt: vyzovy i
perspektivnye resheniya [Trusted Artificial Intelligence: Challenges and Promising Solutions],
Doklady Rossiyskoy akademii nauk. Matematika, informatika, protsessy upravleniya [Reports of the
Russian Academy of Sciences. Mathematics, informatics, control processes], 2022, Vol. 508, No. 1,
pp. 13-18. DOI: 10.31857/S2686954322070207. EDN CVIVCS
