Recent Developments of Artificial Intelligence Attacks and Challenges
S. Kannadhasan, Racshitha Nagarajan, K. Chandramohan, R. Prabhu · Apple Academic Press eBooks · 2025
Department of Electronics and Communication Engineering, Cheran College of Engineering, Coimbatore, Tamil Nadu, India 2Department of Electrical and Electronics Engineering, Gnanamani College of Technology, Namakkal, Tamil Nadu, India 3Department of Computer Science and Engineering, Gnanamani College of Technology, Namakkal, Tamil Nadu, India 4Department of Electronics and Communication Engineering, Gnanamani College of Technology, Namakkal, Tamil Nadu, India Artificial intelligence (AI) has exploded in popularity in recent years, with applications in a broad variety of sectors. AI translates the flow of data into valuable information in a variety of disciplines, including smart governance, smart buildings, smart transportation, smart grids, and smart “everything.” These AI systems are important in the realm of cybersecurity because they gather vast volumes of data and swiftly filter it to detect potentially dangerous patterns and anomalous behavior. As a result, much has been written about AI’s accomplishments, but less has been written about AI’s risks. The bad uses of AI are changing the landscape of possible dangers to a broad variety of positive applications. The possibility of malevolent AI, in 72 particular, might jeopardize more sophisticated systems, such as smart cyberphysical systems, that have yet to be fully explored. Smart cyber-physical systems (sCPSs) are increasingly linked cyber-physical systems that use technologies like the Internet of Things (IoT), artificial intelligence (AI), wireless sensor networks (WSN), and cloud computing to deliver a broad variety of novel services and applications. Because sCPS is so interwoven, a single weakness, such as a butterfly’s wing flap, might trigger a tornado. When one component of a system fails, the whole system collapses with tremendous consequences, similar to the “butterfly effect.” As a result of growing interconnectedness and autonomy, the frequency of cyber assaults has grown. The use of artificial intelligence (AI) for malevolent purposes has increased the reach of cyber threats, enabling larger-scale and more effective assaults. In order to execute more effective operations, cybercriminals have begun to strengthen their techniques by including IoT intrusions, malware, ransomware, and AI into their attacks. Because of their interconnection and intelligence, everyone is in danger when the attacks are carried out. In light of this, despite the fact that research into the use of artificial intelligence (AI) to fight cyberattacks has been ongoing for many years, there is still a significant degree of ambiguity about how to prevent AI from being used as a destructive weapon.