Lighthouse AdvisorySLED AI Adoption Intelligence

Strategic PartnersIssue 062 records

Red Hat

A patch-and-provenance day for Red Hat AI. Red Hat shipped Critical Red Hat AI Inference 3.5.3 builds on Nov. 3 covering 60 CVEs, including vLLM code-execution, SSRF and denial-of-service flaws; Red Hat's own data still lists the 'fast' channel images as Affected. RHEL AI also added a packaged GLM-5.3-Flash model from Z.ai, formerly Zhipu AI, whose entities have been on the Commerce Entity List since January 2025, so public buyers need sizing and provenance review. No SLED deployment news or independent evaluation surfaced.

Evidence records
2
Cross-source patterns
0
Evidence classes
2 vendor claim
Outcomes
1 cautionary1 mixed
Source freshness
2 new this fortnight
Research completed
2026-11-05

Choose a role to see its takeaway beside every record in the ledger.

Patterns across the evidence

Synthesis · Lighthouse Advisory interpretation

No pattern claimed

The evidence in this edition did not support a cross-source pattern. Each record below stands on its own.

Evidence ledger

Full record · every source keeps its link and limitations

2 records
  1. Vendor claimCautionaryNew this fortnight

    Red Hat AI Inference 3.5.3 ships Critical fixes for 60 CVEs, including vLLM code-execution and SSRF flaws, while 'fast' images remain Affected

    Vendor record: on Nov. 3, 2026 Red Hat issued five Critical advisories for Red Hat AI Inference 3.5.3 covering CUDA, CPU, ROCm, Spyre and Gaudi (Tech Preview) builds, each listing the same 60 CVEs. The vLLM fixes include remote code execution through a crafted code_revision when trust-remote-code is enabled (CVE-2026-114053), SSRF through the X-Request-Id header (CVE-2026-116176), unsafe image-format handling (CVE-2026-115915) and denial-of-service flaws. Red Hat's security data, checked the evening of Nov. 4, still lists the rhaii-fast vLLM images as Affected, and marks some older rhaiis images 'Will not fix' for two DoS flaws. An independent advisory entry confirms the ReDoS gap stayed open in one backend after an earlier fix. Interpretation: inference endpoints exposed to campus or agency users should move to 3.5.3 now.

    Red Hat Product SecurityGlobal product advisory; applies to any US SLED deployment of Red Hat AI Inference 3.5RHSA-2026:94702, 94709, 94717, 94721 and 94738 issued Nov. 3, 2026; underlying CVEs public Aug.-Oct. 2026

    Why it matters, evidence and limitations
    Why it matters
    State agencies, universities and research computing centers that self-host open models on Red Hat AI Inference or vLLM behind chat assistants, document tools or student-facing services face network-reachable flaws, some unauthenticated. Self-hosting is often chosen for data residency, so patch ownership sits with the agency or campus, not a SaaS provider.
    Evidence and measured results
    Read RHSA-2026:94738 and RHSA-2026:94702 in full and checked titles and CVE counts for 94709, 94717 and 94721; Red Hat security data JSON for CVE-2026-114053, CVE-2026-116176, CVE-2026-73556 and CVE-2026-93592; GitLab advisory entry for CVE-2026-73556.
    Limitations and uncertainty
    Vendor advisories are terse; no exploitation evidence found (no vLLM entries in CISA KEV as checked Nov. 4); package states observed once; severity differs between Red Hat and third-party scoring.
    RHSA-2026:94738 - Red Hat AI Inference 3.5.3 (CUDA) (Critical) Role takeaways & implementation
  2. Vendor claimMixedNew this fortnight

    RHEL AI adds packaged GLM-5.3-Flash, a Z.ai open-weight model, raising sizing and model-provenance questions for public buyers

    Vendor record: on Nov. 3, 2026 Red Hat issued two RHEL AI enhancement advisories that ship GLM-5.3-Flash as a supported runtime image (amd64 only) and a separate model-car image for several architectures. The Red Hat-hosted model card shows the model comes from Z.ai's repository under an MIT license. Z.ai describes it as 320B total and 18B active parameters. A trade analysis notes the advisories give no accelerator, memory or throughput guidance. Independent context: Z.ai is the former Zhipu AI, and the Commerce Department added Zhipu entities to the Entity List in January 2025. Interpretation: a capable open model is now one registry pull away for RHEL AI users, but public buyers must handle sizing and provenance review first.

    Red Hat (RHEL AI)Global product release; model developer based in China; applies to US SLED RHEL AI customersRHEA-2026:94502 and RHEA-2026:94504 issued Nov. 3, 2026

    Why it matters, evidence and limitations
    Why it matters
    States, universities and K-12 or local IT teams that run RHEL AI may see this model in the supported catalog. Many public bodies have foreign-vendor, supply-chain or AI-use policies that call for review of models from China-based developers, even when weights are open and run on-premises. Research universities also have export-control offices that may want to weigh in.
    Evidence and measured results
    Read RHEA-2026:94502 (and checked RHEA-2026:94504), the upstreambeat analysis, the RedHatAI/GLM-5.3-Flash model card, the Jan. 16, 2025 Federal Register Entity List rule and the Z.ai Wikipedia entry linking the brand to Zhipu.
    Limitations and uncertainty
    No Red Hat sizing or evaluation data; developer benchmarks unverified; legal effect of Entity List on model use is not addressed by any source opened; trade analysis is unbylined.
    RHEA-2026:94502 - Red Hat Enterprise Linux AI (GLM-5.3-Flash runtime) Role takeaways & implementation

How to read this edition

Source findings, measured results and limitations come from the cited publications. Patterns, operating questions, role takeaways and implementation considerations are Lighthouse Advisory interpretation, stated as questions to validate locally rather than guaranteed outcomes. Vendor and operator claims are labeled as claims. Full research method.

Vendor claim
A supplier-provided assertion that has not been upgraded to independent evidence.