Module Directory

This module examines the intersection of artificial intelligence and cyber security, focusing both on the use of AI techniques to enhance security and on the protection of AI systems against attack.

Students explore how machine learning can be applied to tasks such as anomaly detection, intrusion detection and security policy verification, while also analysing threats specific to AI systems including adversarial attacks, data poisoning and model extraction.

The module develops a critical understanding of the security, robustness and trustworthiness of intelligent systems. Through applied laboratories and analytical coursework, students evaluate the effectiveness, limitations and risks of AI-based security solutions and design strategies for building resilient and secure AI-driven systems.

The aims of this module are:

• To provide students with an understanding of how artificial intelligence techniques can be applied to cyber security challenges.


• To examine the vulnerabilities and attack surfaces specific to AI and machine learning systems.


• To enable students to critically evaluate the robustness, reliability and trustworthiness of AI-based security solutions.


• To develop students’ ability to design and assess secure and resilient AI systems.


• To foster awareness of ethical, regulatory and societal implications of AI in security-critical contexts.

By the end of the module, students will be expected to:

1. Critically evaluate the effectiveness of AI techniques in addressing cyber security problems such as anomaly detection and threat analysis.


2. Analyse attack vectors against AI systems, including adversarial examples, data poisoning and model manipulation.


3. Design mitigation strategies to enhance the robustness and security of AI-based systems.


4. Assess the reliability, limitations and risks associated with deploying AI in security-critical environments.


5. Integrate AI and cyber security principles to propose secure, trustworthy and ethically responsible intelligent systems.

Indicative syllabus

• Foundations of machine learning relevant to cyber security


• AI for anomaly detection and intrusion detection


• Behavioural analytics and security monitoring


• AI for policy verification and compliance analysis


• Adversarial machine learning and evasion attacks


• Data poisoning and model integrity attacks


• Model extraction and inference attacks


• Robustness and resilience techniques in ML systems


• Secure model deployment and monitoring


• Explainability and trust in AI systems


• Ethical, legal and regulatory considerations in AI security


• Environmental and computational cost of large-scale AI systems

This module will be delivered via:

• Ten 2-hour lectures
• Ten 2-hour laboratory sessions

Lectures introduce theoretical foundations, research developments and applied case studies. Laboratory sessions provide hands-on experience in implementing AI-based detection systems, experimenting with adversarial techniques and evaluating system robustness. Students are expected to engage with academic literature and technical documentation as part of independent study.

All learning materials will be available through the virtual learning environment, including lecture recordings, lab instructions and supplementary reading.

Inclusive learning is supported through structured practical guidance, progressive laboratory exercises, accessible digital resources and opportunities for formative feedback. Alternative arrangements and reasonable adjustments will be implemented in line with University policy where required.

Laboratory exercises involving adversarial techniques will be conducted in controlled environments with clear ethical guidance to ensure responsible engagement with security topics. CSEE have an existing “Cyber security environment” that has been validated by DITS as providing the suitable controls to operate the laboratories safely.