Juice jacking is a well-known cyber-attack used to attack Universal Serial Bus (USB)-enabled devices such as mobiles, tablets, and laptops. It generally utilizes the charging port of a given device; then, whenever someone connects a given device to the system using this port, the hackers obtain all their personal information or may upload some malware onto the device. Therefore, it is necessary to detect and prevent these kinds of attacks.

Business travelers now have access to public USB power charging stations at airports, hotels, and other places to which they travel or stay [1]. In Android OS and iOS, this attack is more appropriate and the smartphone’s display can be exposed through a standard Micro-USB [2,3] connected using the Mobile High-Definition Link (MHL) standard or the iPhone’s lightning connector.

However, even without the instant injection of malware, a leakage to a rogue kiosk might cause a continual security threat [4]. An address book, images, music, and SMS are just a few of the items that may be accessible once the device is associated with a computer. Both data and power transfer can be accomplished using USB connectors [2,5]. A decade ago, security researchers worked out how to exploit USB connections, which a user might think are solely used to transfer power, to hide and transport secret data payloads, as cellphones grew more prevalent [6,7].

Markus, President of Aries security, and his fellow researchers Joseph Mlodzianowski and Robert Rowley, built the charging kiosk. They made charging stations more attractive with a variety of charging cables. When no device is connected, the charging station displays a blue image with the words “Free cell phone charging kiosk” and whenever any device is connected a red warning sign shows and a message: “Be careful and should not trust public kiosks” [8,9]. The wall of sheep is an event held at Defcon, which has allowed public access to juice jacking kiosks every year since 2011. This can raise awareness among the public [8,10]. In addition, juice jacking can be used by hackers to inject malicious code onto the devices and obtain information on those devices.

Regarding device connectivity, the Internet of Things (IoT) has offered the world a higher level of accessibility, integrity, availability, scalability, secrecy, and interoperability. IoTs, on the other hand, are vulnerable to security threats due to a mixture of various attack surfaces and their newness, resulting in a lack of security standardization and criteria. Attackers can use a wide range of cyberattacks on IoTs, depending on which aspect of the system they are targeting and what they expect to achieve from the attack [11]. As a result, a significant amount of research has been dedicated to building secure IoT devices. Recently, Artificial Intelligence (AI)-enabled approaches have been extensively utilized to implement secure IoT devices and networks. Typically, AI models identify anomalous activity, which helps to predict a given attack. In IoTs, cyber-attackers always have an advantage because they only need to uncover one vulnerability, whereas cybersecurity specialists must secure several targets [12]. Recently, various supervised learning models, such as decision trees, linear regression, machine learning, support vector machines, and neural networks, have been employed in IoT cybersecurity applications to predict threats.