title
Please take a moment to fill out this form. We will get back to you as soon as possible.
All fields marked with an asterisk (*) are mandatory.
Cybersecurity for AI and Agentic Systems
Duration
3 Days
Course
AI-SEC-ADV-2026
Available Formats
Classroom Training, Online Training
Course Description
Overview
- Course goals
- Provide an advanced, practitioner-focused security blueprint for modern AI systems—especially GenAI/LLM and Agentic AI.
- Equip participants, especially technical managers and leads, to enforce, fund, and operationalize security practices across development teams.
- Address the reality that developers often prioritize features and delivery speed over security, and require structured guidance, incentives, and accountability to adopt secure practices.
- This course covers advanced AI-specific security threats, structured threat modeling for GenAI and Agentic AI, robust guardrails, prompt injection and jailbreak analysis, secure data pipelines and deployments, zero-trust frameworks for autonomous agents, and governance and auditing aligned with regulations and standards.
- Special emphasis is placed on:
- Manager-driven security adoption
- Aligning incentives between delivery speed and security
- Embedding security into development workflows without blocking feature delivery
- Turning security expectations into standards, gates, and repeatable engineering practice
- Applying all concepts to a real-world system: FreeEed (open-source eDiscovery platform with AI capabilities)
Format
- Lectures, discussion, case studies, and hands-on labs
- Includes discussion of offensive AI security techniques strictly for educational and defensive purposes, in authorized environments only
- Includes practical strategies for driving secure behavior in real development teams
- Uses FreeEed as a continuous case study across modules
Lab environment
- Hands-on exercises must be conducted in authorized, controlled environments, adhering to ethical and legal guidelines.
- A modern laptop and reliable internet access, plus any client-specific access requirements for lab tooling.
Objectives
- Upon completing this course, participants will be able to:
- Recognize why developers deprioritize security and how to correct this through process, leadership, and incentives.
- Introduce and enforce secure practices in AI teams without bringing delivery to a halt.
- Identify and mitigate advanced AI-specific security threats.
- Develop structured threat models for GenAI and Agentic AI systems.
- Design robust guardrails for LLM-based applications.
- Analyze prompt injection and jailbreak attack techniques.
- Architect secure AI data pipelines and model deployments.
- Apply secure-by-design principles to AI system architecture.
- Implement zero-trust frameworks for autonomous AI agents.
- Establish monitoring, auditing, and governance strategies for AI systems.
Audience
- Technical Managers and Engineering Leaders
- AI/ML Engineers and Architects
- Security Engineers supporting AI deployments
- DevSecOps Professionals
- Technical Leads implementing AI solutions
- Security Researchers and Consultants
Prerequisites
- Foundational understanding of Machine Learning and Large Language Models (LLMs)
- Familiarity with API security and authentication
- General cybersecurity knowledge
- Working knowledge of software development principles
Topics
Day 1: Breaking the Illusion — How AI Systems Get Compromised
- Module 1: Prompt Injection and Jailbreaks
- Learning objectives
- Experience firsthand how easily AI systems can be manipulated.
- Understand why traditional secure coding assumptions fail with LLMs.
- See why teams that focus only on features can expose the business to immediate risk.
- Topics
- Live demonstrations of prompt injection against RAG systems, assistants, and tool-using applications
- System prompt extraction and leakage
- Data exfiltration through model responses
- Tool abuse in agentic workflows
- Direct, indirect, multi-turn, and cross-agent injection
- Why developers miss these risks
- Immediate mitigations
- Lab: Prompt-hacking-and-defenses — perform real prompt injection attacks and implement layered defenses
- Lab: Social-Engineering-and-AI — explore how users and attackers manipulate AI systems through natural language
- Lab: FreeEed — Prompt Injection in eDiscovery — attack a real-world RAG-based system (FreeEed AI Advisor), extract sensitive information, and test guardrails
- Module 2: AI Threat Modeling — Thinking Like an Attacker
- Learning objectives
- Develop threat models specific to AI systems.
- Make threat modeling part of delivery, not an optional exercise.
- Topics
- Why traditional threat modeling is not enough for AI
- Threat modeling frameworks for AI
- Attack surface analysis
- Manager responsibility
- Lab: Threat-Modeling-for-AI — build a full threat model for a GenAI / RAG system
- Lab: FreeEed — Threat Modeling Case Study — model threats across ingestion, indexing, AI querying, and export workflows
- Module 3: OWASP Top 10 for LLM Applications
- Learning objectives
- Apply standardized risk frameworks to AI systems.
- Translate AI security risks into checklists and engineering controls.
- Topics
- Deep dive into the OWASP Top 10 for LLM applications
- Mapping risks to real architectures
- Enforcement strategies
- Lab: GenAI-Top-10 — map vulnerabilities in a real system to OWASP categories
- Lab: FreeEed — OWASP Mapping — identify OWASP Top 10 risks in FreeEed architecture and workflows
- Module 4: Agentic AI Top Risks and Failure Modes
- Learning objectives
- Understand the unique risks of autonomous and semi-autonomous AI systems.
- Topics
- Agentic risk categories
- AI scheming and self-preservation concerns
- Multi-agent risks
- Management controls
- Lab: OpenClawSec — analyze and secure an agentic AI system with tool access
- Module 5: GenAI Security Architecture
- Learning objectives — design secure architectural patterns for GenAI applications.
- Topics — LLM application security architecture; secure design patterns; making architecture non-optional
- Lab: Introduction to AI security — build a baseline secure GenAI application
- Lab: FreeEed — Secure Architecture Review — review and improve FreeEed AI architecture (RAG + services + APIs)
- Module 6: AI Guardrails and Safety Systems
- Learning objectives — design, implement, and monitor AI guardrails.
- Topics — guardrail architecture patterns; content safety and moderation; implementation frameworks; making guardrails stick
- Lab: Guardrails — implement guardrails and evaluate effectiveness
- Lab: FreeEed — Guardrails Hardening — add and tune guardrails for real legal/document workflows
- Module 7: Securing Agentic AI Systems
- Learning objectives — implement security controls for autonomous AI agents.
- Topics — secure agent architecture; zero-trust for AI agents; developer reality
- Module 8: Secure AI Data Pipelines and MLOps
- Learning objectives — secure the AI lifecycle from development through deployment.
- Topics — data pipeline security; model security; MLOps security; organizational controls
- Lab: PenAI — simulate attacks on AI pipelines and defenses
- Lab: FreeEed — Pipeline Security — analyze ingestion, OCR, indexing, and storage for vulnerabilities
- Module 9: API Security for AI Systems
- Learning objectives — protect AI APIs, tools, and integrations.
- Topics — REST API security; AI-specific API attacks; operational protections
- Module 10: Advanced Defense and Red Teaming
- Learning objectives — implement multi-layered protection strategies.
- Topics — defense-in-depth architecture; adversarial testing; making security continuous
- Lab: HarmBench — automated red teaming and benchmarking of AI defenses
- Lab: FreeEed — Red Team Exercise — perform end-to-end attack simulation on a realistic AI system
- Module 11: AI Security Governance and Compliance
- Learning objectives — establish governance frameworks aligned with regulations and internal controls.
- Topics — regulatory landscape; governance frameworks; manager playbook
- Module 12: Future Trends and Emerging Threats
- Topics — multimodal attack vectors; autonomous AI malware; AI-powered security operations; long-term agent risks
- Enterprise multi-agent RAG architectures
- OWASP and MITRE ATLAS mapping
- Guardrails implementation strategies
- Incident response simulations
- Manager-led security transformation examples
- Practical patterns for getting feature-driven teams to adopt secure behavior
- End-to-end FreeEed security blueprint
- Evaluation — final assessment, case study participation, class participation
- Requirements — minimum attendance, successful completion of assessments, completion of module exercises
- Certificate Awarded — Advanced AI Security Professional Certificate
- This course includes discussion of offensive AI security techniques strictly for educational and defensive purposes. All exercises must be conducted in authorized, controlled environments. Participants must adhere to ethical and legal guidelines at all times.
Self-Paced Training Info
Learn at your own pace with anytime, anywhere training
- Same in-demand topics as instructor-led public and private classes.
- Standalone learning or supplemental reinforcement.
- e-Learning content varies by course and technology.
- View the Self-Paced version of this outline and what is included in the SPVC course.
- Learn more about e-Learning
Course Added To Shopping Cart
bla
bla
bla
bla
bla
bla
Self-Paced Training Terms & Conditions
??spvc-wbt-warning??
Exam Terms & Conditions
??exam-warning??
??group-training-form-area??
??how-can-we-help-you-area??
??personalized-form-area??
??request-quote-area??
Sorry, there are no classes that meet your criteria.
Please contact us to schedule a class.
STOP! Before You Leave
Save 0% on this course!
Take advantage of our online-only offer & save 0% on any course !
Promo Code skip0 will be applied to your registration
Nothing yet
Purchase Information
??elearning-coursenumber?? ??coursename??
View Cart
title
Please take a moment to fill out this form. We will get back to you as soon as possible.
All fields marked with an asterisk (*) are mandatory.