Trustworthy AI for Autonomous Cyber Defense: Robustness, Reasoning and Real-Time Assurance

Dear Colleagues,

The rapid evolution of critical infrastructure, robotics, intelligent transportation, industrial control, and mission-critical IoT has led to autonomous defense systems relying increasingly on AI that learns and makes decisions in real time. While such capabilities enhance early detection and accelerate response, adversaries now systematically target the learning pipeline itself, through data poisoning, evasion, backdoors, and model extraction. In cyber–physical contexts, the tight coupling between software and the physical world further elevates the requirements for latency control, safety, and interpretable decision-making.

Given this background, this Special Issue will focus on trustworthy AI for autonomous cyber defense, advancing methods and systems that deliver robustness, verifiability, interpretability, and real-time assurance. We invite high-quality original research and surveys on defenses against adversarial examples and poisoning; provable robustness; AI-based intrusion and threat detection; reinforcement-learning-based active defense and moving target defense; privacy-preserving and federated learning; digital-twin-driven evaluation methodologies; and edge deployment under stringent resource constraints. Submissions should, where possible, report measurable and auditable guarantees—such as robustness bounds, latency budgets, and safety constraints—and provide reproducible artifacts (datasets, code, models, or simulators).

We particularly encourage contributions that bridge algorithmic advances with systems practice, including rigorous evaluations of realistic CPS/IoT/edge testbeds and high-fidelity simulations. By integrating progress across machine learning, cybersecurity, and systems engineering, this Special Issue aims to inform the design of autonomous defensive AI that remains auditable, dependable, and aligned with human operators, thereby strengthening the resilience of next-generation critical infrastructure.

Dr. Fan Liang
Dr. Xing Liu
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. AI is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.