A Systematic Review of Artificial Intelligence (AI)-Enabled Cybercrime and Self-Directed Attack Systems
DS Journal of Artificial Intelligence and Robotics · 31.12.2025 · 3 (4) · 1-43
DOI: 10.59232/air-v3i4p101 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift DS Journal of Artificial Intelligence and Robotics, online erschienen am 31.12.2025. Verfasst von Taban Habibu und 6 weiteren Autor:innen.
Abstract (Originalsprache)
AI has been rapidly transforming how cybercriminals operate and how they attack. The use of AI will greatly affect the scope and level of sophistication that cybercriminals can achieve while executing attacks against organizations. The present work focuses on the development and growth of autonomous systems of attack that are developed and built around AI technology, the ability of AI to learn, reason, and adapt independently, and how the use of AI by cybercriminals can enable them to utilize AI at every stage of the cyber kill chain – Reconnaissance to Command and Control. Through an examination of 46 published studies from 2020 to 2025, the current analysis illustrates the speed, scale, stealth, and ingenuity with which adversaries utilize AI to execute cyberattacks across all phases of the cyber kill chain, as well as the advancement of autonomous offensive tools that the authors refer to in the literature as Recursive Decision Making and Adaptive Evasion as well as large-scale automated attacks. The conclusions of this paper show that traditional means of defence will be increasingly ineffective against future attacks developed and executed through AI by virtue of the inability of defence to detect AI through the gaps and issues associated with current defence systems, including model bias, resource limitations, and lack of transparency in the operation of defence systems. The results of the current study indicate that the shift from human initiated cybercriminal activity to machine initiated and self-learning adversarial systems continues, consequently indicating that an immediate focus must be placed on expanding the proactive and adaptive defence measures needed to deal with a new generation of cybercriminals as well as the enhanced governance and resilience needed to counter the increasing threat posed by these new technologies. Additionally, the present work offers a detailed overview of the current state of knowledge in this domain; identifies the gaps that exist in cybersecurity practices relative to the adoption and use of AI; and discusses how the results of the present study may impact researchers, stakeholders, and the public policy arena as all confront the growing risks posed to both individuals and businesses alike by these emerging technologies.
Zitieren
APA
Habibu, T., Odoch, T., Obudra, C., Afimani, D., Kadabara, B., Ovoni, F. X., & Kasule, B. (2025). A Systematic Review of Artificial Intelligence (AI)-Enabled Cybercrime and Self-Directed Attack Systems. DS Journal of Artificial Intelligence and Robotics, 3(4), 1-43. https://doi.org/10.59232/air-v3i4p101
Vancouver
Habibu T, Odoch T, Obudra C, Afimani D, Kadabara B, Ovoni FX, et al. A Systematic Review of Artificial Intelligence (AI)-Enabled Cybercrime and Self-Directed Attack Systems. DS Journal of Artificial Intelligence and Robotics. 2025;3(4):1-43. doi:10.59232/air-v3i4p101
BibTeX
@article{Habibu2025,
author = {Habibu, Taban and Odoch, Tonny and Obudra, Ceaser and Afimani, Deogratious and Kadabara, Boniface and Ovoni, Francis Xavier and Kasule, Benard},
title = {A Systematic Review of Artificial Intelligence (AI)-Enabled Cybercrime and Self-Directed Attack Systems},
journal = {DS Journal of Artificial Intelligence and Robotics},
year = {2025},
volume = {3},
number = {4},
pages = {1--43},
doi = {10.59232/air-v3i4p101}
}
EUVIMED