Aims and Scope

With the rapid proliferation of artificial intelligence across all sectors of society, ensuring the security, safety, and trustworthiness of AI systems has become a critical global priority. AI and Security Convergence (AISC) is an international, peer-reviewed, interdisciplinary journal that provides in-depth coverage of the latest advances in the convergence of AI and security.

AISC considers original research that focuses on the security and safety challenges arising from the deployment of AI, as well as the use of AI to enhance security solutions. The scope of the journal covers the entire spectrum of AI and security convergence, including (but not limited to):

Data Privacy and Protection: Advanced methodologies for data integrity, provenance tracking, and privacy enhancement in large-scale data mining and analytics.

Large Language Model (LLM) Security: Research on the robustness, safety alignment, and adversarial vulnerabilities of foundation models and generative AI, including prompt injection attacks, backdoor defenses, and hallucination mitigation.

Trustworthy Machine Learning: Studies on privacy-preserving techniques such as Federated Learning, Secure Multi-Party Computation (SMC), and Differential Privacy, ensuring data confidentiality during collaborative model training.

Autonomous Agents and Embodied AI: Security and safety challenges in intelligent agents and embodied systems, focusing on value alignment, behavioral control, and physical-world safety guarantees.

Adversarial Machine Learning: Theoretical and practical aspects of attack and defense mechanisms in deep learning, including data poisoning, model extraction, and adversarial training.

By integrating AI with security engineering, cryptography, and systems safety, the scope of AISC aims to foster the development of resilient, accountable, and human-centric intelligent systems for real-world applications.