Review of AI in Cybersecurity

Status:

The paper has been submitted to a conference, and we are awaiting feedback

Awards:

We have presented the faindings of this paper at the UNERD conference at the University of Toronto. Our work was awarded the "best presentation for the data analytics & AI presentation".

Abstract:

Recent advancements in artificial intelligence (AI) have transformed the cybersecurity landscape. AI powered tools have empowered adversaries to execute sophisticated attacks that exploit system vulnerabilities to inflict severe harm on government institutions, personal data, and financial systems. As a result, interest in developing defense mechanisms against these powerful malicious tools is rising. However, there is a lack of research on how these tools affect the different stages of the Cyber Kill Chain. This study examines how attackers leverage AI at every stage of the chain and explores how defenders can mitigate adverse behavior at each step. We review papers published between 2013 and 2023 from the Web of Science and Google Scholar databases, querying by specific keywords. Then a set of adverse tools and strategies explored in recent research is compiled to build a cybersecurity toolbox which aims to help defenders understand the newest technologies. Furthermore, this research presents emerging defender tools used at every stage of the cyber kill chain. Additionally, this study provides the reader with a comprehensive visual review of the current cybersecurity literature that focuses on AI. Our findings indicate that these malicious tools are most advantageous during the initial stages of the Cyber Kill Chain, where emerging tools do not significantly benefit defenders. The insights provided by this study are valuable for cybersecurity specialists who aim to develop robust defense approaches, an understanding of novel adversarial strategies, and to possibly adapt the Lockheed Martin Cyber Kill Chain framework to address the changing threat landscape.