From the Red Hat edition of November 6, 2026
Checkable AI: Red Hat OpenShift AI security controls assessment (42 findings, last checked 4 Nov 2026)
Checkable AI · Independent AI controls research · Global product assessment (US/EU/AU frameworks referenced)
- Publisher
- Checkable AI
- Original publication
- 2026-11-04
- Source retrieved
- 2026-11-07
- Event date
- 2026-11-04
What happened
Checkable AI’s OpenShift AI stack check (last checked 4 November 2026) reports 42 documentation-based control findings across identity, authorisation, runtime, mediation, observability, agentic MLSecOps and governance. Multiple high-interest gaps are marked Not addressed, including agent identity tied to an owner/release record, tamper-proof administrative audit beyond kube-apiserver config changes, runtime assurance/attestation of inference containers, and SBOM/provenance for MCP servers. The research notes AgentRuntime/OpenShell and MCP gateway capabilities as Developer/Technology Preview and states findings reflect public vendor docs, not a live penetration test. Newly relevant alongside the Nov. 5 OpenShift AI pipeline CVE batch because it frames control debt around the same expanding agent/MCP surface.
Why it matters
Gives SLED architects an independent checklist when justifying OpenShift AI for regulated workloads: which controls are documented GA versus Preview/not addressed. Does not replace agency ATO testing.
Evidence and measured results
Page states 42 findings, last checked 4 November 2026; area tallies; detailed Not addressed notes on Administrative Audit (no tamper-proof record; MLflow/OGX logs not tamper-proof), Runtime Assurance, Agent ID, MCP SBOM/provenance; scope covers OpenShift AI 3.5, Red Hat AI Inference, OGX, Connectivity Link MCP gateway.
Limitations and uncertainty
Documentation review, not a live deployment test. Last checked 4 Nov—one day before this edition’s pipeline CVE wave, so those CVEs are not integrated into the check. Independent site; verify against current Red Hat release notes before contractual use.
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
- Buyers lack an independent view of which OpenShift AI AI-security controls are documented GA versus missing or Preview.
- Stakeholders
- CISO
- ATO/authorizing official
- AI platform owner
- Discovery questions
- Which controls are must-have for your ATO?
- Are agent/MCP features in scope this year?
- Do you require tamper-evident AI audit?
- Value hypothesis
- A controls-gap workshop can focus spend on compensating controls that matter for the agency’s authorization boundary.
- Potential engagement
- Two-week OpenShift AI control crosswalk using Checkable findings and Red Hat release notes.
- Unsupported claims
- Checkable is a certification
- Live exploitability of every Not addressed item
- OpenShift AI is unsuitable for all SLED
Pre-sales engineering
Role takeaway
- Fit
- OpenShift AI 3.x estates evaluating agent/MCP/OGX controls for regulated use.
- Architecture and integration
- Add external immutable logging, signing/attestation, MCP gateway policies alongside product features.
- Prerequisites
- Current release notes
- ATO control catalog
- network policy capability
- Constraints
- Docs-only evidence
- Preview features unsupported in some disconnected modes
- Security
- Gaps concentrate in audit integrity, runtime assurance and MCP supply chain.
- Proof of value
- Completed gap matrix
- Compensating control designs accepted by security
Delivery
Role takeaway
- Work
- Control matrix, compensating designs, Preview feature gates.
- Dependencies
- Security architecture approval
- logging platform
- Ownership
- Security architecture with platform engineering
- Skills and adoption
- Teams learn to operate Preview vs GA feature policy.
- Governance checkpoints
- ATO evidence review
- Preview enablement CAB
- Acceptance criteria
- Signed gap matrix
- Documented bans/allow for Preview MCP/agents
- Risks
- Over-blocking low-risk labs
- Stale Checkable snapshot vs new CVEs
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?
Plan compensating controls outside the product for audit immutability, runtime attestation and MCP supply-chain provenance if those are ATO requirements.
Governance
Who approves, reviews and stays accountable for outcomes?
Treat Preview agent/MCP features as non-production for high-assurance SLED until support and control evidence mature; map each required control to vendor doc status.
Security and privacy
What data, permissions and controls need testing?
Highlights missing tamper-evident AI audit and runtime assurance—material for continuous monitoring plans even when cluster audit exists.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Not a primary focus; Preview agent tooling implies limited supported paths for broader workforce use.
Procurement
What should contracts, pricing and exit terms secure?
Use as a question bank in RFP/ATO evidence requests—ask vendors to show GA controls for audit immutability, runtime assurance and MCP provenance.
Operating model
Which teams own the service once it runs?
Security and platform teams must own compensating controls where Checkable marks Not addressed.
Publication history
- 2026-11-06Red Hat · Issue 085 resources
Stable resource ID: checkableai-openshift-ai-security-controls-assessment-20261104