Advancing cybersecurity and privacy with artificial intelligence: current trends and future research directions

Area of Science:

• Cybersecurity and Privacy
• Artificial Intelligence Applications

Background:

• Escalating cyber threats require innovative cybersecurity and privacy strategies.
• Artificial Intelligence (AI) offers advanced capabilities for intrusion detection, malware classification, and privacy preservation.
• A comprehensive synthesis of AI's role in cybersecurity and privacy is lacking.

Purpose of the Study:

• To synthesize the existing literature on AI applications in cybersecurity and privacy.
• To identify research gaps and guide future progress in the field.
• To analyze global research trends and priorities.

Main Methods:

• Systematic literature review using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework.
• Analysis of over 9,350 publications from 2004 to 2023.
• BERTopic modeling to identify 14 key themes in AI-driven cybersecurity, validated through algorithmic and expert evaluation.

Main Results:

• AI applications are concentrated in intrusion detection, malware classification, federated learning, IoT security, UAV systems, and DDoS mitigation.
• Emerging fields include adversarial machine learning, blockchain, and deep learning.
• Key contributors are the US, India, UK, and China, indicating diverse research priorities.

Conclusions:

• AI's adaptability and scalability are crucial for addressing evolving cyber threats.
• Challenges include computational demands, adversarial vulnerabilities, and ethical concerns.
• Future research should focus on trustworthy AI, standardization, privacy legislation, quantum machine learning, explainable AI, and humanized AI.