Artificial Intelligence in Cybersecurity

Abstract

Artifcial intelligence (AI) and machine learning (ML) have grown rapidly in recent years, and their applications in practice can be seen in many felds, ranging from facial recognition to image analysis. Recent developments in Artificial intelligence have a vast transformative potential for both cybersecurity defenders and cybercriminals. Anti-malware solutions adopt intelligent techniques to detect and prevent threats to the digital space. In contrast, cybercriminals are aware of the new prospects too and likely to adapt AI techniques to their operations.

This thesis presents advances made so far in the field of applying AI techniques in cybersecurity for combating against cyber threats, to demonstrate how this promising technology can be a useful tool for detection and prevention of cyberattacks. Furthermore, the research examines how transnational criminal organizations and cybercriminals may leverage developing AI technology to conduct more sophisticated criminal activities. Next, the research outlines the possible dynamic new kind of malware, called X-Ware and X-sWarm, which simulates the swarm system behaviour and integrates the neural network to operate more efficiently as a background for the forthcoming anti-malware solution. This research proposes how to record and visualize the behaviour of these type of malware when it propagates through the file system, computer network (virus process is known) or by observed data analysis (virus process is not known and we observe only the data from the system). Finally, a paradigm of an anti-malware solution, named Multi agent antivirus system has been proposed in the thesis that gives the insight to develop a more robust, adaptive and flexible defence system.