AI Cyber Defense / Protecting AI systems used by defenders and the business
AI Cyber Defense: Protect model and plugin artifacts through provenance
Use approved artifact sources and pinned revisions, and review extension permissions independently from model quality. Stage updates with behavior and access-boundary checks.
Environment and inputs: Enterprise AI assistants, security copilots, retrieval systems and application-controlled tool executors; framework-independent design.
Scenario and goal
A tampered extension or model artifact can change an assistant's behavior without a visible prompt change.
Why it matters
This path concerns attacks against AI-enabled systems used by the defender or business. It complements defenses against attackers who use AI as a tool. Treat model inputs, retrieved evidence and generated plans as untrusted proposals at execution boundaries. A model may assist analysis, but the application remains responsible for tenant isolation, authorization, resource budgets and the integrity of approvals and receipts.
Evidence to collect
Inventory artifact origin, version, checksum, configuration and the permissions granted to each extension.
Validate with an authorized exercise
Change a harmless test artifact checksum or extension permission in a fixture; verify release review is required before the changed artifact can enter the approved runtime.
Rollout and checks
Use synthetic evidence and mocked write tools for adversarial evaluation. Version the model, prompts, tool contracts and retrieval configuration together. Keep high-impact permissions outside the model and retain an operator takeover path. When an evaluation fails, correct the enforceable boundary and rerun the same fixture rather than relying only on a stronger wording of the prompt.
Scope and limits
This guide proposes a defensive test; it is not a tested production design. Check platform support and local requirements. References explain the mechanisms but do not certify the design. Behavior alone does not prove AI involvement.