{"resourceId":"cisco-ai-transit-sd-wan-neocloud-sovereign-ai-20261102","versions":[{"version":"external-6613747638b73ac38fe29b9c3efaafd3ec6228af8634ac55ae7cc1083f52e454","resource":{"id":"cisco-ai-transit-sd-wan-neocloud-sovereign-ai-20261102","title":"Cisco introduces AI Transit to extend existing SD-WAN policy to NeoClouds, private/sovereign AI clouds and AI factories","organization":"Cisco","sector":"Networking vendor; relevant to SLED agencies and universities using Cisco SD-WAN and external GPU capacity","geography":"Global","publishedAt":"Cisco Blogs, Nov. 2, 2026","publicationDate":"2026-11-02","eventDate":null,"sourceName":"Cisco Blogs","sourceLabel":"Cisco AI Transit connects AI environments to your SD-WAN","sourceUrl":"https://blogs.cisco.com/security/cisco-ai-transit-connects-ai-environments-to-your-sd-wan","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["infrastructure","data-security","governance-procurement"],"finding":"Cisco describes AI Transit, which lets existing Cisco SD-WAN customers bring AI workloads in NeoClouds, private and sovereign AI clouds and enterprise AI factories onto their fabric, deployed as physical or virtual appliances (as-a-service virtual router in participating NeoClouds). It names Armada, Cirrascale and Latitude as providers and highlights geographic intent for controlling where AI traffic travels. No availability date, pricing, performance data or customer deployment is given; the FAQ tells architects to confirm supported environments and controls.","sledRelevance":"Moderate and conditional. Research universities and state agencies buying GPU capacity from specialized clouds, or with data-residency rules, may want one policy model for AI traffic. Only relevant where Cisco SD-WAN is already deployed.","evidence":"Vendor blog announcement; no independent evaluation, benchmarks or SLED customers. No independent coverage found on publication day.","architectureImplications":"Extends SD-WAN overlay and segmentation to AI environments; avoids a separate network domain per AI provider, if supported in the chosen provider.","governanceImplications":"Geographic intent may help document data-residency and traffic-path controls for state AI policies; verify before relying on it.","securityPrivacyImplications":"Consistent segmentation and inspection for AI/agentic traffic are claimed; capabilities must be validated against documentation.","caveats":"Vendor claim; product status, supported providers and pricing unstated; listed providers are not typical SLED contract vehicles.","streamIds":["cisco"],"roles":{"sales":"Sales framing: a discovery opener, not a proven outcome. For SLED customers already on Cisco SD-WAN who are buying or piloting GPU capacity outside hyperscalers, AI Transit is Cisco's answer to keeping one policy model. Ask where AI workloads run now and whether traffic-path controls matter for residency rules. Do not quote availability, pricing or performance; the post provides none, and named NeoClouds may not be on state contracts.","engineering":"Engineering framing: confirm support before designing. AI Transit builds on Cloud OnRamp and runs as a physical or virtual appliance, with a virtual router as-a-service in participating NeoClouds. Validate which providers are supported, how AI and agentic traffic is classified by application policy, and whether segmentation, inspection and geographic intent behave as described. Compare against existing Cloud OnRamp or direct interconnect designs, and test latency and throughput for real inference traffic.","delivery":"Delivery framing: if a customer proceeds, scope a small design and lab validation for one AI environment. Ownership remains with the customer network team, with the AI platform team as stakeholder. Acceptance criteria should be proposals tied to validation, such as policy parity with existing sites and measured latency within agreed bounds. Risks include provider support gaps, contract-vehicle limits and another dependency on vendor roadmap."},"roleDetails":{"sales":{"customerProblem":"SLED organizations adding external GPU capacity risk creating separate network and security domains.","stakeholders":["Network architect","CISO","Research computing director","Procurement"],"discoveryQuestions":["Where do AI workloads run today and where will they run next year?","Do residency rules require controlling traffic paths?","Is Cisco SD-WAN the current WAN fabric?"],"valueHypothesis":"Bounded: one policy model may reduce design effort for new AI environments.","potentialEngagement":"A design workshop and lab validation for one AI environment.","unsupportedClaims":["Availability dates or pricing","Performance or latency gains","Compliance with any specific residency law"]},"engineering":{"fit":"Only for existing Cisco SD-WAN estates using NeoClouds, sovereign or private AI clouds.","architectureIntegration":"Builds on Cloud OnRamp; physical/virtual appliance or as-a-service virtual router.","prerequisites":["Cisco SD-WAN fabric","Supported AI provider","Policy for AI traffic classification"],"constraints":["Provider support limited to participating NeoClouds","Contract vehicles may not include named providers","No published performance data"],"security":"Validate segmentation, inspection and geographic intent controls against documentation.","proofOfValue":["Lab test of policy parity","Measure latency/throughput for inference traffic","Confirm geographic path enforcement"]},"delivery":{"work":"Design, lab validation and pilot connection for one AI environment.","dependencies":["Provider support","Licensing","Research/AI team requirements"],"ownership":"Customer network team.","skillsAdoption":"Network staff need familiarity with AI workload traffic patterns.","governanceCheckpoints":["Architecture review","Security review of new transit path"],"acceptanceCriteria":["Proposed: policy parity verified in lab","Proposed: latency within agreed bounds"],"risks":["Unsupported provider","Roadmap slippage","Vendor lock-in"]}},"retrievedAt":"2026-11-03T04:05:34Z","enrichedAt":"2026-11-03T04:07:53Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Little workforce effect; network staff need AI traffic awareness.","procurementImplications":"New providers may lack SLED contract vehicles; check cooperative contracts.","operatingModelImplications":"Keeps network team as single policy owner for distributed AI.","sourceVerification":{"openedUrl":"https://blogs.cisco.com/security/cisco-ai-transit-connects-ai-environments-to-your-sd-wan","referenceExcerpt":"AI Transit can be deployed as a physical or virtual appliance, depending on the environment.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}