AI-Enhanced Cybersecurity Solutions for Protecting U.S. Aerospace Manufacturing Infrastructure: Techniques and Real-World Applications

Abstract

The aerospace manufacturing infrastructure of the U.S. has been a target of cyberattacks by nation-states and hackers attempting to poach its intellectual property. Cyberattacks are conducted to compromise the confidentiality, integrity, or availability of computer systems and networks. They are exploited via malware, distributed denial of service attacks, exploiting vulnerabilities, and human errors. Cyberattacks can have severe consequences on the attacked systems, creating economic vulnerabilities for the affected entities. A federal sUAS acquisition directive was enacted to strengthen the cybersecurity posture of contractors that provide systems with software components that could be exploited with malware during the electronic supply chain. Current cybersecurity solutions rely on the use of rules-based systems and threat intelligence feeds to detect cyberattacks. To augment such defensible architectures in U.S. aerospace manufacturing infrastructure of interest, AI-enhanced network intrusion detection systems are reviewed for potential application. The research stage of the supply chain security architecture is detailed and a literature review is conducted. Four AI-enhanced cybersecurity solutions, supported by published proof-of-concept and research prototypes, are examined. Each implementation would operate independently or could be seamlessly integrated into a layered approach to covering attack vectors associated with reconnaissance, exploitation, and installation levels of the cyber kill chain [1] , [2].