Skip to Content
View all events

AI Attack Surface: The 10 Most Critical Threats to ML and AI Systems

Published by Pearson

Advanced content levelAdvanced

Practical Defense Strategies for AI Cyber Threats

  • Attack-Centric Approach: This training focuses on the 10 most critical AI attack vectors, providing a practical framework for understanding and mitigating specific threats, unlike general AI security overviews.
  • Hands-on and Practical: It goes beyond theory with step-by-step instructions, real-world examples, case studies, and exercises, enabling learners to apply their knowledge immediately.
  • Structured Learning Path: Offers a clear, organized curriculum covering the entire AI lifecycle, from data poisoning to model extraction and evasion techniques, providing a comprehensive understanding of AI security.

This live course, “AI Attack Surface: The 10 Most Critical Threats to ML and AI Systems”, offers practical, step-by-step guidance to identify, analyze, and mitigate AI-specific attack vectors.

It lays the groundwork, introducing post-quantum cryptography and its relevance to AI, then delves into 5 key attack vectors targeting AI systems. It explores 5 additional attack vectors and pivots to defensive strategies, covering post-quantum cryptographic techniques and best practices for building resilient AI. The training concludes with a hands-on activity, solidifying learned concepts.

What you’ll learn and how you can apply it

  • Analyze AI-specific attack vectors.
  • Implement effective defense strategies against adversarial attacks.
  • Conduct comprehensive AI security risk assessments.

This live event is for you because...

  • You are a cybersecurity professional (security analyst, penetration tester, threat intelligence expert) aiming to understand and address the unique security challenges posed by AI systems.
  • You are an AI researcher or developer (machine learning engineer) committed to building secure and resilient AI models and deployments.
  • You are a regulatory or compliance officer seeking to navigate the evolving landscape of AI governance and security standards.
  • You are an academic or AI ethics scholar interested in the practical security implications of AI technologies.
  • Whether you're a beginner eager to grasp the fundamentals of AI security or an experienced professional seeking to master advanced threat mitigation techniques, this course offers essential knowledge and practical tools.

Prerequisites

  • This course is designed for professionals with a basic understanding of AI and cybersecurity.

Course Set-up

  • Attendees will be required to access the platform using a computer with a stable internet connection. No specific software or setup is required beyond this, as all resources will be provided directly through the interactive platform.

Recommended Preparation

  • Watch: AI Security and Responsible AI Practices by Omar Santos and Petar Radanliev: AI Security and Responsible AI Practices
  • Attend: Agentic AI and Cybersecurity Risks by Petar Radanliev: Agentic AI and Cybersecurity Risks
  • Attend: Algorithmic Red Teaming in Cybersecurity Risks by Petar Radanliev: Algorithmic Red Teaming in Cybersecurity

Recommended Follow-up

Schedule

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

Introduction to AI Security (30 mins)

  • Why AI Security Matters
  • The Anatomy of AI Cyber-Attacks

The 10 Most Dangerous AI Cyber-Attacks (30 mins)

  • Adversarial Perturbations
  • Data Poisoning Attacks

Break (10 mins)

Model Extraction and LLM Threats (30 mins)

  • Model Extraction Attacks
  • LLM-Specific Threats

Social Engineering and Federated Learning Attacks (30 mins)

  • Social Engineering and Deepfakes
  • Attacks on Federated and Reinforcement Learning

Break (10 minutes)

Attacks on AI-Powered Security Systems (30 mins)

  • Attacking AI-Powered Security Systems

Defending AI Systems (30 mins)

  • AI Security Best Practices

Interactive Capstone Activity (20 mins)

  • Practical threat mapping exercise or AI defense checklist

Final Wrap-Up and Q&A (20 mins)

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

Generative AI