What you are evaluating
The research-map entry focuses on AI Model Security within Prisma AIRS. AI Runtime Firewall, AI Runtime API, red teaming and agent protection are separate documented areas; verify their licenses and deployment paths rather than assuming platform branding enables them.
A useful evaluation context
A team adopting external or internally trained models can evaluate supply-chain checks before allowing artifacts into an application.
Documented capabilities
The vendor describes these capabilities in the linked sources. Availability depends on the product edition and supported environment.
- Model scanning examines supported model files and embedded code before deployment.
- API integration supports incorporating artifact assessment into build and deployment workflows.
- The broader platform documents runtime inspection, red teaming and posture functions with separate operational scopes.
Where it fits in the work
- Record the origin and version of a harmless test model artifact.
- Inspect its assessment output and trace a finding to the exact artifact or dependency.
- Keep promotion approval separate from runtime tool permissions, then test the deployed application independently.
APPLY THE IDEA / ILLUSTRATIVE EXERCISE
Make the outcome observable.
Compare a known test artifact with a changed dependency manifest in a non-production build.
Evidence to look for
The review preserves artifact identity, the finding and promotion decision; it makes no claim that the scan certifies runtime behavior.
Use synthetic data and an authorized test environment. Agree the scope and recovery steps before enabling enforcement.
Questions for your evaluation
- Which formats and artifact sizes does this scanner support?
- Where do scanning and evidence storage occur for this deployment?
- Which agent permissions remain outside the model scan?