Overview

As artificial intelligence (AI) becomes increasingly integrated into modern enterprise systems, it plays a dual role — as both a powerful defense tool and a potential attack vector. Organizations are rapidly adopting AI to strengthen security through intelligent threat detection, automated response, and predictive analytics. However, the same technologies also enable AI-driven cyberattacks, such as deepfake-based social engineering, automated phishing, adaptive malware, and adversarial attacks targeting AI models themselves.

The fusion of AI and cybersecurity thus creates a new digital battlefield where enterprises must defend against intelligent adversaries while leveraging AI for protection. Enterprises face growing challenges in securing data pipelines, protecting machine learning models from manipulation, and ensuring that automated defenses remain trustworthy, explainable, and resilient against evolving threats.

This track explores the latest research, systems, and practices in AI-driven cybersecurity for enterprises — focusing on both attack and defense perspectives. It aims to foster collaboration between researchers, practitioners, and industry professionals to build robust and adaptive security frameworks for the AI era.

Topics include but are not limited to

• AI-enhanced intrusion detection and prevention systems
• Machine learning models for threat intelligence and malware analysis
• Adversarial machine learning and model robustness
• AI-assisted phishing and social engineering attacks
• Deepfake detection and defense mechanisms
• Secure model training and data privacy in AI systems
• Automated security incident response using AI
• Ethical and explainable AI for enterprise security decisions
• AI-based risk assessment and anomaly detection in enterprise networks
• AI-powered red teaming and penetration testing tools
• Data poisoning and model evasion in enterprise AI environments
• AI governance and compliance frameworks for cybersecurity

Important Dates

Submission Guidelines

Submissions must be original and unpublished. All papers should be submitted via the online platform CMT and follow the IEEE formatting guidelines provided here: IEEE Template.

Papers must be in English and should not exceed 6 pages, including figures and references. Authors are required to include their names and affiliations.

Chairs

Mahdi Khosravy, Cross Labs, Kyoto. Japan, & Kyoto College Graduate Studies in Informatics, Japan

Kazuaki Nakamura, Tokyo University of Science, Japan

Aleksandar Shurbevski, KCG, Japan

Joe Austerweil, Chiba Institute of Technology, Japan

Program Committee Members