Explore UCD

UCD Home >
Blue padlock labeled AI regulation on digital microchip background, concept for artificial intelligence data security, policy enforcement, and compliance
overlay image

AI Security: Offensive, Defensive, and Operational Best Practices

Overview

Module Code COMP41870
Module Title AI Security: Offensive, Defensive, and Operational Best Practices
Subject Area Computer Science
Credits 10
NFQ 9
EFQ 7
Start Date 19th January 2026
Duration 12 Weeks
Time Flexible (1-2 lectures per week)
Mode of Delivery Online with in-person end of term exam
Course Leader Dr. Nhien An Le Khac
Fee

€1,369

Application Deadline

15th August 2025

Register your interest: (opens in a new window)microcredentials@ucd.ie 

This “AI Security” module equips students with offensive and defensive techniques to secure AI architectures, generative models, and AI-driven applications. Spanning background fundamentals, practical labs, and policy frameworks, the course ensures graduates can identify and mitigate security flaws while complying with emerging regulations and operating AI systems securely at scale.

All applicants must be current employees of a law enforcement organisation (LE) working in an investigative role. LE includes any organisation that has responsibility for the enforcement of national or local legislation such as:

  • Local, National, or Federal Police Forces
  • Government law enforcement agencies
  • Immigration & Border Control, Revenue & Customs
  • Defence forces including military police
  • International Police organisations (Europol, Interpol, etc)
  • Regulatory Enforcement agencies etc
  • Public prosecutors / Attorney General's office

Although it is preferred that applicants hold a primary degree in computer science or a relevant subject, this is not a compulsory requirement and applicants working in the field of digital forensic investigations in law enforcement for more than 2 years will be eligible for the course without a third level degree. 

Module Dependencies/ Prerequisites: 

The students are required to have the background of:

  • Introductory Cybersecurity: Understanding of threat modeling, network security basics (required).
  • Python programming: Proficiency in coding, debugging, and using Python-based (required).

On successful completion of this module the learner will be able to:

  1. Explain foundational AI and cybersecurity concepts relevant to AI pipelines.
  2. Identify and exploit vulnerabilities in AI models using recognized adversarial attack methods.
  3. Devise and implement robust defenses (secure training, monitoring, adversarial mitigation) within modern practices.
  4. Integrate governance, ethics, and compliance considerations (e.g., EU AI Act, bias) into AI security strategies.
  5. Evaluate new threats and propose forward-looking solutions to secure AI systems against future attack trends.

Indicative Content:

  1. AI Security Landscape
    • AI Foundations - Architectures, Pipelines, and Machine Learning Frameworks
    • Cybersecurity Essentials for AI – Attack Surfaces & Defensive Principles
  1. Offensive Security
    • Threat Modeling & Risk Assessment for AI
    • Adversarial Attacks (CV & NLP)
    • Generative AI Exploits & Prompt Injection
    • Generative Agent Security Fundamentals
    • Data Poisoning & Model Backdoors
    • Privacy Attacks & Inference Risks
  1. Defense
    • Robust Training & Adversarial Defenses
    • AI Security Policy Design & Implementation
    • Monitoring, Logging, & Anomaly Detection
    • Governance, Compliance, & Ethical AI
  1. Operational Best Practices & Emerging Topics
    • Incident Response & AI Forensics
    • Case Studies & Industry Insights
    • Future Trends

Benefits will vary based on student background and employment both current and prospective. 

  • For law enforcement officers, having a qualification to professionalise their training has the additional advantage of adding credibility to their testimony as expert witnesses. 
  • Career development possibilities may be enhanced. 
  • This course offers a theoretically grounded, practical curriculum for understanding, detecting, and mitigating these threats, with lectures providing essential background in both AI fundamentals and cybersecurity concepts—so learners of varied experience can confidently tackle advanced topics later in the module.

This course is delivered online with an end of term, in-person exam. 

Lectures are pre-recorded and provided online via a virtual online learning environment. 

Exercises, reading, assignments and projects are conducted online and in our virtual labs. 

Students are required to come to Dublin or another UCD exam centre (Netherlands, North America and Hong Kong) to complete in-person examinations. 

All modules are examined in May in one of our five exam centres internationally. Please note that an exam centre venue will be subject to availability and demand.

Student Effort Hours:

Autonomous Student Learning (98 Hours)

Lectures (24 Hours)

Practical Learning (48 Hours)

Online Learning (30 Hours)

Total Hours over 12 weeks: (200 Hours)

Approaches to Teaching and Learning:

  • Active/task-based learning; 
  • Lectures; 
  • Critical writing; 
  • Reflective learning; 
  • Lab work; 
  • Problem-based learning; 
  • Case-based learning.

All applicants must be current employees of a law enforcement organisation (LE) working in an investigative role. Applicants do not have to be a sworn officer. LE includes any organisation that has responsibility for the enforcement of national or local legislation such as:

  • Local, National, or Federal Police Forces
  • Government law enforcement agencies
  • Immigration & Border Control, Revenue & Customs
  • Defence forces including military police
  • International Police organisations (Europol, Interpol, etc)
  • Regulatory Enforcement agencies etc
  • Public prosecutors / Attorney General's office

Although it is preferred that applicants hold a primary degree in computer science or a relevant subject, this is not a compulsory requirement and applicants working in the field of digital forensic investigations in law enforcement for more than 2 years will be eligible for the course without a third level degree. 

Each applicant will be assessed on a case by case basis. Students are also required to fulfil UCD’s English Language Requirements (see http://www.ucd.ie/international/study-at-ucd-global/ucdenglishlanguagerequirements/)

Assessment Strategy:

  • Exam (In-person): (70%)
  • Assignment: (30%) Two assignments (10% and 20% in Week 4 and Week 8 respectively)

This micro-credential is stackable with other micro-credentials, COMP47510 and COMP41880 for a 30 credits Graduate Certificate award in Forensic Computing and Cybercrime Investigation (F007).

  • Leadership in Security COMP47800
  • Applied Cryptography COMP47890
  • Risk Assessment and Standards COMP47900
  • Secure Software Engineering COMP47910
  • Information Security COMP47920
  • Cybersecurity Law LAW42160
  • Incident Response COMP47870
  • Network Security COMP47880
  • Malware Analysis COMP47810
  • Ethical Hacking COMP47860

On successful completion of this micro-credential, you will receive credits as per the European Credit and Transfer System. These credits are recognised by the awarding institution as credits aligned to learning completed at postgraduate level.