Cybersecurity Labs
AI SECURITY
AI systems become are increasingly
becoming targets for sophisticated cyber-
attacks due to their complex and evolving
nature. We must ensure ethical, reliable, and
unbiased performance to prevent misuse
and enhance overall system resilience
Try free AI lab
https://crac-learning.com/ admin@crac-learning.com +91 74289 73398
AI SECURITY
This training offers an introduction
to cybersecurity fundamentals in
artificial intelligence and machine
learning. Participants will engage in
hands-on labs using DefHawk's
vulnerable AI application platform.
We will cover key AI/ML security threats,
such as Prompt Injection, Poison Training
Data, Model Asset Compromise, Insecure
Output Handling, and Model Theft.
Participants will learn secure development
practices, including secure coding, model
validation, and effective deployment strategies.
The course also explores tools and techniques to
mitigate cybersecurity threats in AI/ML systems and best
practices for data security, such as encryption, secure
storage, and data anonymization.
Key Learning Outcomes
U nderstanding AI/ML- S
pecific Cybersecurity Threats
Gain in-depth knowledge of unique security risks such as Prompt
Injection, Poison Training Data, Model Asset Compromise, and Model Theft.
Li ve Demonstration of S ecure AI/ML Practices
v H ’
v
De elop practical skills through labs on Def awk s intentionally
ulnerable AI platform.
Implementing Data S ecurity B est Practices
U se techniques like encryption, secure storage, and data anonymisation
to ensure compliance and robust security.
Core Module Topics
Introduction
Introduction to the course objectives, structure, and expectations
Overview of AI and cybersecurity fields and their intersection
Overview of basic web application components.
Basic scripting using python
Familiarity with Networking and Systems.
Understanding Cybersecurity controls with Data Handling and
Data security.
Predictive AI vs. Gen AI
What is predictive AI?
What is Gen AI?
Application of Predictive AI
Application of Gen AI
Introduction to LLMs
AI Techniques and Applications in Industry
Cybersecurity in AI
Real life security issues in AI
Statistics of Security issues
Common mistakes in AI
Common attacks on AI
Introduction to cybersecurity
Definitions, goals, and the importance of cybersecurity
Cybersecurity Threats and Vulnerabilities
Cybersecurity best practices
Common threats (e.g., malware, phishing, DoS attacks) and
vulnerabilities
Basic principles like the CIA triad (Confidentiality, Integrity,
Availability)
LAB on log4shell CVE
Introduction to cybersecurity attacks
Client-side attacks
XSS
CSRF
Server-Side attacks
File Upload
Server-side template injection
Understanding attacks in detail
Case studies from industry
General web application Attacks
Command Injection
Cross Site Scripting
Arbitrary file actions
Local file inclusion
Sensitive information disclosure
Labs on AI application attacks listed above
Understanding attacks in detail
Prompt injection
Format corruption
Poison training data
Model asset compromise
Adversarial attacks on the model
Hardcoded secrets and configuration flaws
Weak access control and authentication
Deployment and infrastructure flaws
Conclusion
Curated By Experts From
Internship-to-placement pipeline
Build the technical foundation and practical experience required
to excel as a Web Security Analyst, Application Security Engineer,
or Penetration Tester.
Multi-level structured learning path
Learn step-by-step approaches to both exploiting and
defending against vulnerabilities in web applications.
Industry-backed curriculum
Learn how to identify and adapt to the latest trends and
techniques in the rapidly evolving field of cybersecurity.
https://crac-learning.com/ admin@crac-learning.com +91 74289 73398
