Build a foundation in offensive AI security by learning how AI systems are designed, where they fail, and how adversaries exploit them, using structured hacking methodologies and globally recognized AI security frameworks.

What You will Learn

• Understand AI and machine learning fundamentals from an offensive security perspective
• Identify AI attack surfaces, threat landscapes, and adversary techniques
  aligned to MITRE ATLAS
• Apply AI system hacking methodologies, frameworks, and risk implications
• Classify AI attack taxonomies and models
• Define offensive AI scoping fundamentals and foundations for securing AI systems
• Provide an overview and mapping of OWASP LLM & ML Top 10 (2025) to AI threats and governance considerations

Learn advanced AI-focused OSINT techniques to identify, enumerate, and analyze AI assets, data pipelines, models, APIs, and attack surfaces, and apply exposure mitigation and hardening strategies to support continuous AI
security monitoring.

What You will Learn

• Apply OSINT tools and techniques to identify and profile AI assets
• Gather intelligence from AI data sources and training pipelines
• Discover and map AI attack surfaces using publicly available intelligence
• Enumerate AI endpoints, services, APIs, and exposed parameters
• Identify and analyze AI models and vector stores from an attacker’s perspective
• Evaluate OSINT exposure and apply hardening controls to reduce risk
• Use AI threat intelligence to support continuous monitoring and defensive
  readiness

Master AI-specific vulnerability assessment and fuzzing techniques to identify, analyze, and mitigate security weaknesses across modern AI systems and applications.

What You will Learn

• Understand core principles of AI vulnerability assessment and threat discovery
• Use tools and techniques for scanning vulnerabilities in AI models, pipelines, and deployments
• Apply practical fuzzing methods tailored for AI systems and model interfaces
• Integrate scanning and fuzzing into AI security workflows for proactive risk
  mitigation

Analyze and exploit LLM trust boundaries using advanced prompt injection, jailbreaking, and output manipulation techniques, while identifying risks related to sensitive data exposure and insecure LLM application design.

What You will Learn

• LLM architecture, trust boundaries, and associated attack vectors
• Execute prompt injection and jailbreaking techniques in real-world LLM applications
• Identify sensitive information disclosure and system prompt leakage risks
• Evaluate improper output handling vulnerabilities and misinformation threats
• Apply advanced prompt-based attack techniques and exploitation strategies
• Implement secure LLM application design principles and defensive controls

Execute and analyze adversarial machine learning, privacy, and model extraction
attacks to assess AI system robustness, trustworthiness, and risk, and apply defensive strategies to mitigate them.

What You will Learn

• Identify core adversarial machine learning attack classes
• Execute practical adversarial input attacks across data modalities
• Apply privacy, inference, and model extraction attack techniques
• Evaluate robustness, trustworthiness, and risk evaluation methods
• Implement defensive strategies for model privacy and resilience

Compromise AI systems through data poisoning and backdoor insertion targeting
training pipelines and model integrity.

What You will Learn

• Understand AI data and training pipeline architecture and threat surfaces
• Execute practical data poisoning techniques and attack scenarios
• Apply backdoor and trojan insertion during model training
• Implement security measures to safeguard data and training pipelines

Analyze and exploit autonomous AI agents and multi-model architectures by targeting excessive agency, cross-LLM interactions, orchestration workflows, and unbounded resource consumption, while understanding defensive strategies to secure agentic systems.

What You will Learn

• Understand agentic AI architecture and attack surface
• Apply excessive agency and autonomy exploitation techniques
• Identify cross-LLM and model-to-model attack vectors
• Asses denial-of-wallet risks and unbounded resource consumption
• Execute attacks targeting AI workflows and orchestration layers
• Implement defensive strategies for securing agentic AI applications

Explore offensive techniques targeting AI infrastructure, system integrations, and third-party dependencies, while learning how to identify, exploit, and harden AI supply chain weaknesses.

What You will Learn

• Understand AI infrastructure components and system integration architectures
• Identify vulnerabilities in AI systems, frameworks, and deployment pipelines
• Analyze abuse of tools, plugins, and APIs in AI-enabled applications
• Assess AI supply chain threats and dependency risks (deep dive)
• Implement hardening strategies for AI infrastructure and supply chains

Apply structured AI security testing and evaluation methodologies to assess
risk, validate controls, and implement hardening best practices across enterprise AI
systems.

What You will Learn

• Understand AI security testing methodologies and evaluation techniques
• Apply red team frameworks for offensive AI assessment
• Identify, validate, and report AI vulnerabilities and risk
• Implement security hardening and mitigation best practices for AI systems

Master AI-specific incident response and forensics, concluding with hands-on
engagement in AI red team activities.

What You will Learn

• Detect and respond to AI-specific security incidents
• Collect and analyze AI logs, telemetry, and digital evidence
• Analyze root causes in post-incident analysis