From the ServiceNow edition of November 5, 2026
Kickstart Your ServiceNow Brazil Release Guide (Part 1 of 4)
ServiceNow (Community article by ServiceNow employee) · Enterprise platform release management; applies to all ServiceNow customers including public sector · Global
- Publisher
- ServiceNow Community (Upgrades and Patching blogs)
- Original publication
- 2026-11-05 (posted ~10 hours before retrieval; marks Brazil GA)
- Source retrieved
- 2026-11-06
- Event date
- 2026-11-05
What happened
Brazil reached General Availability on 2026-11-05. The kickstart guide frames Brazil as moving from working alongside AI to supervising autonomous AI, opening governed actions to AI agents through MCP servers. It warns MCP servers are multiplying across ITSM, ITOM, SPM, HRSD and Process Intelligence (CMDB and Service Mapping in Controlled Availability), and that Platform Owners must decide which AI clients may connect before enabling them. Legacy LLM models are no longer selectable for new AI skills, agents or agentic workflows. Catalog-role ACL removals on request tables and other access changes require pre-upgrade testing; many production upgrades are expected in December.
Why it matters
Why-now: Brazil GA day. Agencies and universities on ServiceNow must plan MCP client allow-lists, AI model migration off legacy LLMs, and request-ACL regressions before December production upgrades. No SLED-specific deployment metrics; guidance is vendor upgrade advice.
Evidence and measured results
Vendor-employee community kickstart post for Brazil GA. Operational claims (MCP availability, legacy LLM restriction, catalog ACL changes) are stated as release facts but not independently verified against full product release notes in this edition.
Limitations and uncertainty
Vendor upgrade guide; full PREPARE impact matrix linked as 'coming soon' in places; not a measured SLED outcome study.
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
- Public-sector ServiceNow customers face a Brazil GA upgrade with expanding MCP/AI surfaces and access changes without a clear readiness plan.
- Stakeholders
- ServiceNow platform owner
- CISO
- Service catalog owner
- Agency CIO
- Discovery questions
- When is your Brazil production upgrade planned?
- Which MCP servers are you entitled to enable?
- Which AI clients could connect today?
- Who still relies only on catalog roles for request access?
- Value hypothesis
- A structured readiness workshop may reduce upgrade regressions and ungoverned MCP exposure.
- Potential engagement
- Brazil upgrade readiness assessment focused on MCP, ACL and AI-model cutover.
- Unsupported claims
- Brazil autonomous AI outcomes are proven in SLED
- All MCP servers are generally available
- Legacy models stop working for existing published assets
Pre-sales engineering
Role takeaway
- Fit
- Fits any ServiceNow customer planning Brazil upgrade who uses or will use AI agents/MCP.
- Architecture and integration
- Product MCP servers (ITSM/ITOM/SPM/HRSD/Process Intelligence; CMDB/Service Mapping CA) plus Otto naming and optional Lux UI.
- Prerequisites
- Target Brazil release notes
- Entitlement check per MCP server
- Sub-production instance
- Constraints
- CA for some MCP servers
- December upgrade congestion
- Deprecations require remapping
- Security
- Decide AI-client allow-lists before enabling MCP; remediate request ACLs; review Vault Console sensitive-data alerts.
- Proof of value
- MCP inventory with allow-list decisions documented
- Request ACL regression tests pass
- New agents not selecting deprecated legacy models
Delivery
Role takeaway
- Work
- Run PREPARE checklist: MCP, ACL, model, Lux and deprecation tests in sub-production.
- Dependencies
- Release notes
- Security approval
- Catalog owner user remediation
- Ownership
- Platform owner with security and catalog co-owners.
- Skills and adoption
- Train admins on MCP controls and Otto naming changes.
- Governance checkpoints
- MCP client approval
- Upgrade go/no-go with test evidence
- Build Agent access decision
- Acceptance criteria
- Proposed: 100% entitled MCP servers have allow-list decisions
- Proposed: zero known catalog-role request access breakages in test
- Risks
- Enabling MCP without client controls
- Catalog ACL lockouts
- Incomplete legacy-model inventory
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?
More product MCP servers become available for external AI clients; CMDB/Service Mapping MCP still Controlled Availability. Build Agent enablement and Lux UI adoption are separate optional tracks.
Governance
Who approves, reviews and stays accountable for outcomes?
Platform Owners need an explicit decision process for which AI clients may connect to each MCP server and who may use Build Agent before users discover it.
Security and privacy
What data, permissions and controls need testing?
Expanding MCP surface increases agent tool risk if clients are enabled without allow-lists; Vault Console default sensitive-data alerts affect work-note handling. Legacy model cutover for new agents needs inventory of existing published assets that may continue on older models.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Upgrade testing load rises for platform, security and service-catalog owners; December upgrade wave compresses skill preparation.
Procurement
What should contracts, pricing and exit terms secure?
Confirm product-tier entitlements for each MCP server before enabling; Controlled Availability items may need separate enrollment.
Operating model
Which teams own the service once it runs?
Shift from ad-hoc AI enablement to supervised autonomous AI with pre-upgrade access/MCP checklists and named owners for test evidence.
Publication history
- 2026-11-05ServiceNow · Issue 063 resources
Stable resource ID: servicenow-brazil-ga-kickstart-mcp-ai-supervision-20261105