Systematic Literature Review of Challenges and AI Contributions in Penetration Testing

Sammanfattning

This systematic literature review aims at understanding the position and application of the AI inchanging the cybersecurity perspective, particularly penetration testing. Penetration testing is animportant process that does involve the testing of the system in order to focus and highlight theweaknesses that hackers may use to gain access into a certain system. This research aims to exploretrends, approaches, and challenges of AI for penetration testing through a systematic review ofliterature. This knowledge is paramount in helping the practitioners, researchers, and policymakers inharnessing AI’s potential in enhancing cybersecurity. It has been examined that AI has the potentialto revolutionize the penetration testing process by improving its automation, effectiveness, andflexibility through assessing vulnerabilities, anticipating attacks and automating tests. Nevertheless,some questions are still open, what constitutes strong enough training data, how to defend againstadversarial AI. It also includes an explanation for what kind of ethical issues can arise from usingsuch models. The research examined that one cannot deny the existence of these issues wheneveraddressing the question of how to increase the cybersecurity level using AI possibilities in usingpenetration testing for enhancement, there are important odds in employing this solution. This callsfor the need to embrace positive results of implementing AI in security practices which on the otherhand pose negative impacts of integrating the technology. Implications have also been evaluated inthe study which include the need for Institute attention to the careful specific AI integration andusage and of plans to identify and manage associated risks and for further exploration of pertinentethical controversies and comparative optimal models that may exist in various disciplines acrossInstitute fields of study.