From the Palo Alto Networks edition of November 5, 2026
Prisma AIRS November 2026 notes: public Kubernetes IP-tag harvesting for runtime firewalls and FP/FN reporting with auto-remediation on API Intercept
Palo Alto Networks · AI runtime security / Kubernetes and API guardrails · Multi-cloud (Amazon EKS, Azure AKS, Google GKE) and API Intercept customers
- Publisher
- Palo Alto Networks TechDocs
- Original publication
- Prisma AIRS documentation, New Features November 2026 (release date November 2026)
- Source retrieved
- 2026-11-06
What happened
Prisma AIRS November 2026 release notes describe two capabilities. First, AI Runtime Security can harvest IP-tag mappings from public and hybrid Kubernetes clusters (EKS, AKS, GKE) into Dynamic Address Groups via Strata Cloud Manager, so Prisma AIRS AI Runtime Firewalls (VM-Series) can enforce policy on ephemeral pod and inference-endpoint labels instead of static CIDRs. Administrators onboard cloud accounts with read permissions, configure an IP-Tag Collector profile and assign the firewall as a redistribution agent. Second, AI Runtime Security API Intercept gains false-positive and false-negative reporting with auto-remediation: operators dispute verdicts from scan logs and AI violation dashboards, triggering exception patterns and local rule adjustments, with reported verdicts also fed into Palo Alto's detection pipeline for model retraining. Documentation only; no measured false-positive reductions or SLED deployments.
Why it matters
Universities and state agencies running containerized AI inference on public cloud Kubernetes, and those already using API Intercept guardrails, get concrete operational features. Interpretation: tag-based policy helps when research workloads churn IPs; FP/FN auto-remediation can cut alert fatigue but needs change control so exception patterns do not quietly weaken protections for FERPA or HIPAA prompts.
Evidence and measured results
Opened and read the full November 2026 new-features page for both feature sections, including deployment steps and auto-remediation description.
Limitations and uncertainty
Vendor documentation; month-level dating only; no quantitative FP/FN results; auto-remediation effects on detection quality not independently tested.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-06; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Ephemeral Kubernetes AI workloads break static IP policies, and noisy API Intercept verdicts create operational friction.
- Stakeholders
- Cloud platform lead
- AI application owner
- Security operations
- Compliance officer
- Discovery questions
- Are inference workloads on public EKS, AKS or GKE?
- Is AI Runtime Firewall (VM-Series) already deployed?
- How are false positives on API Intercept handled today?
- Who may authorize exception patterns?
- Value hypothesis
- Interpretation: tag-based runtime policy plus a controlled FP/FN loop could stabilize guardrails for dynamic AI workloads.
- Potential engagement
- A non-production IP-tag harvesting pilot plus an FP dispute runbook review.
- Unsupported claims
- Measured FP rate reduction
- Automatic coverage of private-only clusters without onboarding
- Compliance certification from these features alone
Pre-sales engineering
Role takeaway
- Fit
- Fits Prisma AIRS runtime customers with public/hybrid Kubernetes AI workloads or API Intercept noise problems.
- Architecture and integration
- Strata Cloud Manager collectors feed Dynamic Address Groups on AI Runtime Firewalls; API Intercept console feeds local exceptions and vendor retraining.
- Prerequisites
- Cloud account onboarding with cluster read permissions
- IP-Tag Collector profile and redistribution agent
- API Intercept console access
- Constraints
- Public/hybrid cluster focus as documented
- Auto-remediation can change local policy quickly
- Month-level release dating only
- Security
- Limit cloud read roles; require review of exception patterns that weaken prompt-injection or DLP rules.
- Proof of value
- Observe DAG update when a labeled pod starts
- File a synthetic FP and inspect created exception
- Confirm policy uses tags rather than static CIDRs
Delivery
Role takeaway
- Work
- Onboard cloud accounts, configure collectors, author tag-based AI Security Profiles, and stand up FP/FN governance.
- Dependencies
- Cloud IAM approvals
- VM-Series runtime firewall capacity
- SecOps change-control process
- Ownership
- Platform networking owns IP-tag pipelines; SecOps owns verdict disputes and exceptions.
- Skills and adoption
- Staff need training on Kubernetes labels and AIRS violation dashboards.
- Governance checkpoints
- Cloud read permissions reviewed
- Exception-pattern dual control enabled
- Monthly audit of FP/FN changes
- Acceptance criteria
- Proposed: test workload protected by tag-based rule within agreed window
- Proposed: FP dispute creates a reviewed exception, not a silent allow
- Proposed: rollback path for auto-remediation documented
- Risks
- Over-privileged cloud collectors
- Exception patterns reopening blocked content classes
- Tag hygiene failures leaving workloads unprotected
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Needs Strata Cloud Manager cloud onboarding, IP-Tag Collector profiles and VM-Series redistribution; API Intercept console becomes a feedback loop into local exceptions and vendor model retraining.
Governance
Who approves, reviews and stays accountable for outcomes?
FP/FN disputes and auto-created exceptions should require dual control or periodic audit in regulated environments.
Security and privacy
What data, permissions and controls need testing?
Cloud read permissions for cluster metadata expand the control-plane attack surface; exception patterns can reopen previously blocked prompt classes.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Security teams need Kubernetes tagging literacy and a workflow for verdict disputes.
Procurement
What should contracts, pricing and exit terms secure?
Confirm AI Runtime Firewall (VM-Series) and API Intercept entitlements separately; features are release-note claims, not new public pricing.
Operating model
Which teams own the service once it runs?
Platform and SecOps must agree who owns IP-tag pipelines and who approves FP/FN exceptions.
Publication history
- 2026-11-05Palo Alto Networks · Issue 076 resources
Stable resource ID: prisma-airs-november-2026-k8s-ip-tag-harvesting-fpfn-auto-remediation