AI Cyber Defense / Protecting AI systems used by defenders and the business
AI Cyber Defense: Partition agent memory by owner and purpose
Store only necessary facts with provenance and explicit scope. Apply access checks and expiry outside the model, and require review before promoting evidence into durable operational guidance.
Environment and inputs: Enterprise AI assistants, security copilots, retrieval systems and application-controlled tool executors; framework-independent design.
Scenario and goal
A misleading fact from one task can influence another user if memory lacks an authorization boundary.
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
Review memory owner, source, expiration, access controls and retrieval context for each stored fact.
Validate with an authorized exercise
Write a synthetic fact in one test user's memory and query from another; verify no cross-user retrieval and confirm an expired fact is omitted or clearly marked stale.
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.