C1 · AI Security Engineering (core)
AI System Architecture and Threat Modeling
Inventory AI systems, map data and authority, and translate threats into testable security requirements.
Proposed
0 of 5 modules accessible. Joining Skool does not unlock planned modules.
At a glance
- Audience
- Security practitioners taking responsibility for AI systems.
- Prerequisites
- API, identity, and data-flow fundamentals
- Scripting
- Planned duration
- 12 hours. Assumes the entry requirements are met. Planning estimate, not a measured completion time. Per module: 2 to 4 hours, plus practice.
- Assessment
- Architecture packet graded against a rubric.
- What is free
- The free foundation module listed on each course page, as it releases, plus community access.
Job tasks this course prepares you for
- Inventory an AI service and its owners
- Trace data and authority
- Turn threats into requirements with test methods
- Communicate residual risk
Learning outcomes
- Produce an inventory containing every seeded model, identity, tool, data store, and accountable owner.
- Draw a data-flow diagram marking every seeded trust boundary and external egress path.
- Write at least six security requirements, each with a test method.
- Distinguish model-behavior constraints from application-enforced authorization in a written design review.
Modules
00System componentsProposed
Topics
- model API, application, retrieval, vector store, orchestration, tools, memory, serving environment.
01Inventory and ownershipProposed
Topics
- models, providers, data sources, identities, tools, versions, business actions, dependencies.
02Trust and authority flowsProposed
Topics
- user to application, retrieval to context, model to tool, agent to agent.
03Threat modeling and requirementsProposed
Topics
- attacker access, assets, impact, control owner, verification method.
04Risk communicationProposed
Topics
- assumptions, uncertainty, residual risk, escalation, acceptance authority.
Labs and deliverables
- Examine a small support assistant. Architecture packet (inventory, data flow, authority map, threat model, prioritized requirements).
Labs run on SIFT, the forensics workstation Rob T. Lee created, on your own machine. Each module states what it needs before you start. Each module confirms its setup requirements and any cloud or API costs before it opens. About the SIFT lab environment.
Assessment
Architecture packet graded against a rubric. A named assessor grades against the published rubric (rubric to be published before release).
Pass criteria
- 80% rubric
- Identify every deliberately seeded high-impact authority path
- No requirement may rely only on the model choosing to obey
Critical failures
- Unauthorized cross-tenant access
- Out-of-scope activity
- Uncontrolled destructive action
- Fabricated evidence
- Exposed credentials
What this course does not teach
- No model training, comprehensive cloud security, or legal interpretation.