Skip to Content
View all events

Algorithmic Red Teaming in Cybersecurity

Published by Pearson

Intermediate content levelIntermediate

Tackle AI vulnerabilities, simulate attacks and implement top defense strategies

  • Gain hands-on experience simulating AI cyber threats and securing AI systems.
  • Identify AI vulnerabilities and assess security risks through red teaming exercises.
  • Defend against adversarial attacks, data poisoning, and model evasion while ensuring compliance.

AI systems are increasingly targeted by cyber threats, with attacks propagating through organizations, supply chains, and broader ecosystems. This course focuses on Algorithmic Red Teaming, equipping professionals with the skills to assess, model, and mitigate AI cybersecurity risks. Participants will explore the impact of AI attacks, learn how adversaries exploit vulnerabilities, and develop proactive defense strategies to safeguard AI-driven infrastructures.

Focusing on threat modeling, adversarial testing, and risk propagation, this course features real-world case studies, hands-on exercises, and cutting-edge research to help organizations build resilience against AI-specific cyber threats. By the end of the course, participants will have the expertise to simulate AI attacks, analyze cascading risks, and implement mitigation strategies to secure AI systems at scale.

This course is designed for cybersecurity professionals, AI engineers, red teamers, and policymakers seeking to understand how AI models are exploited and how to build cyber-resilient AI systems.

What you’ll learn and how you can apply it

  • Conduct algorithmic red teaming by simulating adversarial attacks on AI models to evaluate their security posture.
  • Identify and exploit vulnerabilities in AI-powered systems, including data poisoning, model inversion, adversarial perturbations, and evasion attacks.
  • Assess and mitigate cascading risks in AI-driven infrastructures, ensuring resilience across organizations, supply chains, and critical sectors.
  • Implement AI security best practices, such as adversarial training, differential privacy, model hardening, and anomaly detection techniques.
  • Use AI threat modeling frameworks to map attack vectors, quantify risks, and develop mitigation strategies for enterprise AI systems.
  • Translate AI security research into actionable defense strategies, enabling organizations to prioritize cybersecurity investments and comply with industry standards.
  • Integrate AI security, governance, and policy to ensure AI systems align with cybersecurity regulations and ethical AI principles.

This live event is for you because...

  • You are a cybersecurity professional, AI researcher, red teamer, risk analyst, or technology enthusiast seeking to develop advanced skills in AI security and adversarial testing.
  • You want to understand how AI systems are attacked and how cyber threats propagate across organizations, supply chains, and ecosystems.
  • You aim to gain hands-on expertise in algorithmic red teaming, including adversarial attack simulation, AI-specific threat modeling, and security risk assessment.
  • You are responsible for securing AI-driven systems and need to implement proactive defense strategies against adversarial ML threats and real-world attack vectors.

Prerequisites

  • This course is designed for professionals from all backgrounds and does not require prior knowledge of AI cybersecurity. This course starts with foundational concepts and progresses to practical applications. Participants should have a basic understanding of technology and a willingness to learn. A computer with a stable internet connection is required for hands-on activities during the course

Course Set-up

  • This live online course is delivered via an interactive platform, enabling real-time engagement with the instructor and peers. Participants will use industry-standard tools and software for hands-on exercises, with step-by-step guidance provided throughout the sessions. Course materials, including slides, code examples, and additional resources, will be shared electronically.

Recommended Preparation

Recommended Follow-up

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Session 1: Introduction to Algorithmic Red Teaming & AI Cybersecurity Risks (50 min)

  • Overview of Algorithmic Red Teaming
  • AI Attack Surface Analysis
  • Case studies of Real-World AI Cyber-attacks
  • Threat Modelling Exercise
  • Using MITRE ATLAS to map attack vectors in different AI models
  • Identifying attack pathways in an enterprise AI system
  • Hands-on demonstration of a simple AI evasion attack

Q&A (5 minutes)

Break (5 minutes)

Session 2: Adversarial Attack Vectors and Risk Propagation (50 min)

  • AI Threat Categories: Evasion, poisoning, model inversion, and backdoor attacks
  • Risk Propagation in AI Systems and Supply Chains
  • Case Study: Deepfake & LLM Security in Misinformation Campaigns
  • AI Supply Chain Risk Analysis Exercise
  • Live Demo: Prompt Injection Attack on an LLM Chatbot
  • AI Security Risk Assessment Exercise

Q&A (5 minutes)

Break (5 minutes)

Session 3: AI Red Teaming Techniques and Security Assessments (50 min)

  • Principles of Algorithmic Red Teaming: Methodologies, ethics, and best practices
  • Security Assessment Frameworks: OWASP AI Security, MITRE ATLAS, and NIST AI Risk Management
  • Introduction to Adversarial Robustness Testing and Model Evaluation
  • Red Teaming Exercise: AI security assessment on a pre-trained NLP model
  • Testing Model Robustness: Adversarial training, input sanitization, and anomaly detection
  • Red Teaming an AI Model: Simulating evasion attacks and testing robustness with AutoAttack or CleverHans

Q&A (5 minutes)

Break (5 minutes)

Session 4: Defensive Strategies for AI Security, Mitigation, and Policy (55 min)

  • Defensive Strategies: Adversarial training, model interpretability, differential privacy
  • Ethical AI Red Teaming: Governance, compliance, and regulatory frameworks
  • Future AI Threats: Adversarial AI, self-replicating malware, nation-state threats
  • Implementing adversarial robustness and evaluating security improvements
  • Differential privacy techniques
  • Red Team vs. Blue Team: AI attack response and countermeasures

Q&A (5 minutes)

Your Instructor

  • Dr. Petar Radanliev

    Dr. Petar Radanliev lectures and supervises postgraduate master’s students’ research dissertations on AI and cybersecurity at the Department of Computer Science, University of Oxford. He is also a Lecturer/Instructor at Pearson and O’Reilly (USA), while conducting research on digital identity system security at the Alan Turing Institute, based at the British Library in London. After completing his PhD in 2013/14, Petar held postdoctoral research appointments at Imperial College London, the University of Cambridge, the Massachusetts Institute of Technology, and the Department of Engineering Science at the University of Oxford, where he remained for seven years before moving to his current position. His work spans artificial intelligence, cybersecurity, post-quantum security, and blockchain security. This research has led to an H-index of 25 (as indexed by Web of Science and Scopus), over 3,700 citations, more than 100 peer-reviewed publications, and four authored books. In recognition of his contributions, Petar has received major funding awards, including a Fulbright Fellowship and the Prince of Wales Innovation Award.

Skill covered

Artificial Intelligence (AI)