Part 2 — For the Developer
Ch. 10 — Risk Classification and System Boundaries
How to define and enforce the boundaries of what your agent system is — and what it is not allowed to become.
Risk classification is done against a defined system and intended purpose, not the model in isolation. Agentic systems can drift beyond that approved scope when their runtime boundaries are not enforced in code.
10.1 Key Takeaways
A static model is relatively easy to classify. It has fixed inputs, fixed outputs, and fixed capabilities. An agent is none of those things.
Platform Agentic
Compliance, governance, and accountability for teams building agentic AI systems.