Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 21 hours
Course Outline
Foundational Frameworks: Threat Modeling for Agentic AI
- Categorizing agentic risks: misuse, privilege escalation, data exfiltration, and supply-chain vulnerabilities
- Defining adversary profiles and attacker capabilities relevant to autonomous agents
- Mapping organizational assets, trust boundaries, and critical control points for agent operations
Governance, Policy, and Risk Management
- Establishing governance frameworks for agentic systems (roles, responsibilities, and approval gates)
- Developing policies: acceptable use, escalation protocols, data handling standards, and audit requirements
- Addressing compliance obligations and securing evidence for governmental audits
Non-Human Identity & Authentication for Agents
- Architecting agent identities: service accounts, JSON Web Tokens (JWTs), and short-lived credentials
- Enforcing least-privilege access patterns and implementing just-in-time credentialing
- Managing the identity lifecycle: rotation, delegation, and revocation strategies
Access Controls, Secrets, and Data Protection
- Implementing fine-grained access control models and capability-based patterns for agents
- Managing secrets, ensuring encryption in transit and at rest, and adhering to data minimization
- Safeguarding sensitive knowledge sources and Personally Identifiable Information (PII) from unauthorized agent access
Observability, Auditing, and Incident Response
- Designing telemetry for agent behavior: intent tracing, command logging, and provenance tracking
- Integrating with Security Information and Event Management (SIEM) systems, setting alert thresholds, and maintaining forensic readiness
- Developing operational runbooks and playbooks for agent-related incident containment
Red-Teaming Agentic Systems
- Planning red-team exercises: defining scope, rules of engagement, and safe failover mechanisms
- Simulating adversarial techniques: prompt injection, tool misuse, chain-of-thought manipulation, and API abuse
- Conducting controlled attacks to measure system exposure and potential impact
Hardening and Mitigations
- Deploying engineering controls: response throttling, capability gating, and sandboxing
- Implementing policy and orchestration controls: approval workflows, human-in-the-loop interventions, and governance hooks
- Applying model and prompt-level defenses: input validation, canonicalization, and output filtering
Operationalizing Safe Agent Deployments
- Adopting deployment patterns: staging, canary releases, and progressive rollouts for agents
- Enforcing change control, testing pipelines, and pre-deployment safety checks
- Facilitating cross-functional governance: integrating security, legal, product, and operations playbooks
Capstone: Red-Team / Blue-Team Exercise
- Executing a simulated red-team attack against a sandboxed agent environment
- Defending, detecting, and remediating as the blue team using implemented controls and telemetry
- Presenting findings, a remediation plan, and proposed policy updates
Summary and Next Steps
Requirements
- Proficient background in security engineering, system administration, or cloud operations
- Familiarity with AI/ML concepts and the behavioral characteristics of large language models (LLMs)
- Experience with Identity and Access Management (IAM) and secure system design principles
Audience
- Security engineers and red-team specialists
- AI operations and platform engineers
- Compliance officers and risk managers
- Engineering leads responsible for agent deployments
Testimonials (1)
inventory and identifying the different risk exposures within AI
