Architecture Lens

Secure generative AI by controlling the context path

Generative AI security is not limited to model access. The context path also includes source approval, retrieval filters, identity-aware permissions, prompt behavior, evaluation, logging, and human review. Each layer needs clear controls before users rely on generated output.

  • Authorize retrieval by user identity and document permissions, not only by application access.
  • Log prompt, retrieval, response, policy decision, and human review events at the right level of detail.
  • Use evaluation sets for privacy, hallucination, source attribution, toxic content, and restricted workflows.
Secure RAG control path
User request

Identity, session, device, and business context.

Policy gate

Allowed action, data scope, and safety rules.

Retrieval

Approved sources, permissions, freshness, and citations.

Review

Evaluation, logging, escalation, and feedback.

Original InSkyto diagram informed by NIST Generative AI Profile and Microsoft Zero Trust principles.

References

NIST Generative AI Profile Microsoft Zero Trust guidance

Delivery Pattern

Treat prompts like application behavior

Prompt rules, grounding instructions, refusal behavior, and escalation paths shape what users can do. They should be versioned, tested, reviewed, and monitored like application logic.

Checklist

Evidence to collect before launch

Before launch, keep evidence for approved sources, prompt versions, evaluation results, access rules, human review criteria, and response monitoring so the team can explain system behavior under audit or incident review.