In recent years, there is a new generation of cyber-attacks based on advances in artificial intelligence (AI) and machine learning (ML) algorithms. It is sophisticated because it takes advantage of the availability of various heterogeneous data from several information sources, such as social networks, cookies, and other Internet user activities. It uses data science conjointly with AI and ML to generate highly targeted and hard-to-detect personalized attacks. Therefore, a new era of cyber-security mechanisms based on AI and ML is being developed to deal with such trends of cyber-attacks. Open-source tools and virtual machines already exist for developing malicious tasks. We will be faced with them more and more in the coming years. AI will remove the human dimension of the attack, making it very difficult to identify the perpetrators. However, the best way to face these new sophisticated cyber-attacks is to use AI and ML-based cyber-security algorithms capable of fighting these attacks using the same methods and post-infiltration data analyses.

By recovering more data than necessary, hackers risk attracting the attention of the security team. Therefore, the AI tools are used to automate the traditional attack process, reduce risk factors, and increase returns. For instance, chat-bots become friends with users of social networks. These bots would have learned beforehand what type of profiles to look for. They would have already interacted with users and could create credible and authentic-looking content. Once chat-bots gained the trust of their victims, hackers could gain valuable intelligence on the target users that would then be exploited to craft compelling targeted phishing attacks for an initial intrusion. ML algorithms would preselect relevant data before extracting them to gain attack capacity and increase the hackers’ return on investment through intrusion. To anticipate such a new evolution of threats, it must now adopt defensive solutions based on AI and ML as a platform to neutralize ongoing cyberattacks, when security experts are unavailable to respond to rapidly spreading threats. This provides targeted and proportionate protection as soon as it detects abnormal activity. It gives cyber-security teams time to focus on what matters while repelling attacks.

The aim of this Special Issue is to attract original research articles and review articles about recent work and advances related to the use of AI and ML algorithms as countermeasures and cyber-security mechanisms against cyber-sophisticated attacks. We hope this Special Issue brings together researchers from academia and the industry in the field.

Potential topics include but are not limited to the following:

• New generation of cyber-attacks based on AI and ML
• AI- and ML-based algorithms for preventing the new cyberattack generation
• AI- and ML-based algorithms for discovering vulnerabilities in wireless networks and Internet
• Applications of AI and ML algorithms as countermeasures for Internet of Things (IoT) cyberattacks
• Network intrusion detection (NDI) using AI and ML algorithms
• Open-source software and potential security concerns
• AI- and ML-assisted malware detection
• AI and ML algorithms detecting and responding to disinformation in social networks
• AI and ML algorithms to track malicious behaviour in social networks
• Ethics, regulation, and privacy issues in the use of AI and ML algorithms in cyber-security