Cybertronium Certified DefenAI Professional

Course Duration5 days
Start Date10 Nov 2025
Delivery MethodFace-to-face

Overview

The Certified DefenAI Professional (CDAIP) course is designed to equip professionals with the knowledge and skills needed to identify and mitigate risks associated with AI exploitation and adversarial AI attacks. The course focuses on protecting AI assets from various threats like data poisoning, model inversion, and adversarial examples. The goal is to help companies safeguard their AI models from both cybercriminals and other AI systems, preventing financial losses, reputational damage, and legal liabilities.

Who Should Attend?

This course is for a wide range of professionals, including

Data Science Analysts, AI Engineers, Developers, and Architects. It also targets security-focused roles like;

Pentesters, Security Analysts, Bug Bounty Hunters, and Red Team members. Participants should have a basic understanding of cybersecurity and AI concepts, though familiarity with machine learning and Python is helpful but not required.

Course Outcome

Upon completing the course, students will be able to understand different attacks on various AI models, including

Large Language Models (LLMs) and Deep Learning Models (DLMs). They will learn how to analyze vulnerabilities, develop mitigation strategies, and design effective defense mechanisms to protect AI modules from attacks. The course also teaches how to identify and understand different exploitation techniques like

model inversion, adversarial examples, and data poisoning.

Course Outline

The course is five days long and is structured as follows:

Day 1: Introduction to AI and Machine Learning concepts, development lifecycle, and ethical principles.
Day 2: Focuses on the AI security landscape, common attack vectors, threat modeling, and best practices.
Day 3: Covers specific attacks on different AI models and data sources, including LLMs, Deep Learning Models, and Tree-Ensemble Models, as well as data poisoning.
Day 4: Explores attacks on AI infrastructure, APIs, and techniques for jailbreaking LLMs.
Day 5: Delves into advanced attack techniques like Membership Inference Attacks, Model Inversion, and Model Extraction, along with defensive strategies and robustness. The final day also includes a course recap and assessment.