Cybersecurity and Artificial Intelligence: Triad-Based Analysis and Attacks Review

This study aims to expand the understanding of Artificial Intelligence (AI) attack scenarios and develop effective protection mechanisms against them. The triadic principle was used to investigate attacks on traditional systems and AI systems, enhance these attacks using AI, and employ AI for cybers...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Cybernetics and information technologies : CIT Ročník 25; číslo 3; s. 156 - 185
Hlavní autoři: Veprytska, Olena, Kharchenko, Vyacheslav, Illiashenko, Oleg
Médium: Journal Article
Jazyk:angličtina
Vydáno: Sofia Sciendo 01.09.2025
De Gruyter Brill Sp. z o.o., Paradigm Publishing Services
Témata:
Artificial Intelligence (AI)
Countermeasures
Cyberattacks
Cybersecurity
Machine Learning (ML)
ISSN:1314-4081, 1311-9702, 1314-4081
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This study aims to expand the understanding of Artificial Intelligence (AI) attack scenarios and develop effective protection mechanisms against them. The triadic principle was used to investigate attacks on traditional systems and AI systems, enhance these attacks using AI, and employ AI for cybersecurity defence. By systematically analysing the interactions between these elements, we create a comprehensive set of attack scenarios and corresponding defensive strategies. Current analysis reveals distinct attack patterns and vulnerabilities associated with traditional and AI-based systems. Effective defence mechanisms and strategies were identified and tailored to various attack scenarios, leveraging AI’s capabilities for improved security measures. The findings provide a structured approach to understanding and mitigating AI-related threats in cybersecurity. By mapping out the roles of AI in both attack and defence, this study offers valuable insights for developing advanced tools and methods to assess system security and enhance countermeasures.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1314-4081
1311-9702
1314-4081
DOI:10.2478/cait-2025-0028