ServiceNow used its recent Knowledge 2026 conference to advance the platform-level argument that enterprise AI will be decided not by model capability alone, but by the infrastructure that governs, orchestrates, and executes AI across the business.

During his keynote, CEO Bill McDermott framed the moment as an inflection point for enterprise software, arguing that the company that controls AI governance and workflow execution controls the enterprise operating layer.

Key Announcements

The centerpiece announcements collectively advance ServiceNow’s bid to serve as the governance and execution layer for agentic AI across every major enterprise function. Each announcement builds on ServiceNow’s foundational strength, shaped by 22 years of encoded enterprise process logic.

AI Control Tower

The AI Control Tower, first introduced at Knowledge 2025 as a visibility and management product, is the architectural spine of Knowledge 2026. ServiceNow expanded it into a five-dimension governance framework: Discover, Govern, Secure, Observe, and Measure.

The scope now extends beyond ServiceNow-native agents to include AI models, agents, and workflows on AWS, Azure, Google Cloud, SAP, Oracle, and Workday. ServiceNow announced 30 new enterprise integrations in total.

All AI Control Tower capabilities are now included across every product tier, enabling every ServiceNow customer to have governance tooling by default.

Key capabilities include:

• Continuous discovery of AI agents as they appear, including shadow AI outside of ServiceNow
• Risk scoring and least-privilege access enforcement for AI agents
• Real-time runtime monitoring with alerting when agents behave unexpectedly
• ROI measurement against governance standards
• Cross-vendor visibility, including integration with Microsoft Agent 365 to surface agents built outside ServiceNow

ServiceNow also deepened its Microsoft partnership to extend AI Control Tower governance across the Microsoft Agent 365 ecosystem and to enable ServiceNow AI Specialists to operate within Microsoft 365 applications, including Outlook, Word, and PowerPoint, with metered usage tracked across both platforms.

Action Fabric

Action Fabric is a new capability that allows any third-party AI agent, including Anthropic Claude, Microsoft Copilot, and others, to securely trigger governed enterprise actions within the ServiceNow platform.

This capability establishes ServiceNow as the execution and governance substrate through which external agents perform real work. Action Fabric provides the connective tissue linking the proliferating universe of AI agents to the process and data infrastructure where enterprise work happens.

Autonomous Workforce Expansion

ServiceNow launched the Autonomous Workforce concept in early 2026 with an IT service desk specialist. Knowledge 2026 expanded the program to every major enterprise function.

The company introduced what it calls AI Specialists, chains of AI agents assigned to specific business roles.

New specialists announced include:

• Case Management: triages customer contacts, resolves issues, stores intent and sentiment context, escalates to humans on request
• CRM Sales: qualifies leads, generates quotes, handles renewals, resolves invoice disputes
• HR: manages employee lifecycle workflows, handles requests across payroll, benefits, and onboarding
• Finance: handles financial reconciliation and approval workflows
• Security and Risk: integrates with Armis and Veza for autonomous threat response (see below)
• Legal: handles contract routing, review workflows, and compliance tasks

Autonomous CRM

ServiceNow entered the CRM market with Autonomous CRM, which covers sales, service, quoting, order fulfillment, invoice disputes, renewals, and customer lifecycle management. The product suite includes Autonomous CRM for Sales, Autonomous CRM for Service, and Configure, Price, Quote (CPQ).

The announcement is a calculated competitive move. Salesforce announced its own ITSM offering aoround the same time, opening a front in both fronts.

ServiceNow’s workflow and process orchestration foundation provides differentiation in execution depth, while Salesforce holds advantages in installed base, data breadth, and CRM ecosystem maturity.

Autonomous Security and Risk (Armis + Veza Integration)

ServiceNow integrated its recent acquisitions of Armis and Veza into a unified Autonomous Security and Risk offering.

Armis contributes real-time visibility of every connected asset across complex enterprise environments. Its intelligence feeds directly into ServiceNow’s CMDB, converting a historically static inventory into a live, contextual view of the attack surface.

Veza governs identity and access.

Together, the platform provides:

• Continuous asset discovery and risk-contextual CMDB updates
• Real-time access governance and least-privilege enforcement
• Faster, context-aware incident response drawing on both asset state and identity data
• Integration with AI Control Tower for agent-level security governance

Project Arc (with NVIDIA)

ServiceNow and NVIDIA announced Project Arc, an enterprise autonomous desktop agent that extends agentic AI into the operating environment. Project Arc uses NVIDIA OpenShell to create a secure sandbox for desktop-level agent execution, and ServiceNow AI Control Tower governs every action.

The design principle is full auditability, with every agent action logged, policy-bound, and reversible.

RaptorDB

ServiceNow highlighted RaptorDB as the high-performance database engine underpinning the entire platform. RaptorDB is designed to support AI processing at enterprise scale, addressing the throughput and latency demands that those agentic workflows place on transactional infrastructure.

Analysis

ServiceNow used Knowledge 2026 to crystallize what it’s been building toward for several years: a platform that governs and executes agentic AI across the enterprise.

ServiceNow’s long history of encoded business process logic, including approval chains, escalation rules, compliance workflows, and data relationships across CMDB, ITSM, HR, and security, provides a foundation that AI-native entrants cannot quickly replicate.

The platform’s process authority gives AI agents the context to act correctly:

• The AI Control Tower’s expansion from product to a platform-default governance capability is a durable move, making governance a baseline expectation for every ServiceNow customer.
• Action Fabric’s openness to third-party agents (Anthropic, Microsoft) is strategically important, signaling that ServiceNow accepts that not all enterprise AI will run on its platform and instead bets on being the governed execution substrate that all agents must pass through.
• The Autonomous CRM entry creates a new revenue surface and a new competitive front; ServiceNow’s workflow depth is a differentiator, but it enters a market where Salesforce holds a significant head start in data breadth and ecosystem development.
• The NVIDIA partnership on Project Arc extends ServiceNow’s reach to desktop-level autonomy, a layer where it has had no prior presence; execution risk here is higher than in its core workflow territory.

Competitive Landscape

The competitive dynamics emerging from the announcements are multidirectional. ServiceNow competes with Salesforce on CRM and workflow automation, with Microsoft on AI orchestration and governance, and, increasingly, with enterprise platform vendors, including SAP and Oracle, as it expands its autonomous function coverage.

Capability	ServiceNow	Salesforce	Microsoft
AI Agent Governance	AI Control Tower: cross-vendor, built into all tiers	Einstein Trust Layer: Salesforce-native scope	Copilot Dashboard: Microsoft ecosystem focus
Agentic Orchestration	AI Specialists across IT, HR, Finance, Security, CRM	Agentforce: sales and service emphasis	Copilot Studio: broad, developer-centric
CRM Automation	Autonomous CRM (new): workflow-native execution	Core strength; established market leader	Dynamics 365: integrated but fragmented
IT Service Management	Core platform; 22-year workflow foundation	ITSM (new challenger; announced K26)	ServiceNow via native integrations
Security & Risk Automation	Autonomous Security & Risk (Armis + Veza)	No comparable autonomous security offering	Security Copilot: advisory, not execution
Third-Party Agent Interop	Action Fabric: Claude, Copilot, and others	Limited external agent ingestion	Agent 365: integrates ServiceNow via partnership
Enterprise Data Foundation	RaptorDB + CMDB	Data Cloud	Microsoft Fabric + Azure AI Foundry

ServiceNow’s clearest differentiation in enterprlies ise process authority and cross-vendor AI governance. No other platform has comparable depth in encoded business process logic across IT, security, HR, and finance.

The AI Control Tower’s cross-vendor scope, including visibility into AWS, Azure, Google Cloud, SAP, and Oracle agent deployments, addresses a gap that platform-native governance tools (Einstein Trust Layer, Copilot Dashboard) structurally cannot fill.

The Salesforce dynamic deserves specific attention. Salesforce announced an ITSM offering at approximately the same time ServiceNow announced Autonomous CRM, indicating both companies have drawn the same conclusion, that the enterprise software market is converging toward full-platform AI execution, and the boundaries between CRM, ITSM, and ERP are dissolving.

Salesforce enters ITSM with limited process depth; ServiceNow enters CRM with limited data breadth. The outcome of this competitive expansion will take several product cycles to resolve.

Final Thoughts

The breadth of the announcements reinforces that ServiceNow is building a horizontal platform rather than a collection of vertical AI applications. The consistency of the governance argument across all product areas reflects a deliberate bet that enterprises will pay a premium for AI they can trust, audit, and control.

The execution challenges are real. The Autonomous Workforce roadmap is ambitious, spanning every major enterprise function within a single year. Autonomous CRM enters a category where Salesforce has structural advantages that workflow depth alone cannot overcome.

Project Arc’s desktop autonomy is architecturally interesting but operationally unproven at the enterprise scale. And the Action Fabric’s openness to third-party agents, while strategically sound, creates integration and support complexity that grows with each new agent type admitted to the platform.

Enterprises evaluating ServiceNow’s agentic capabilities should prioritize the AI Control Tower as an immediate, tangible action item, regardless of broader platform decisions. Its default inclusion across all tiers removes the cost barrier, and its ability to discover shadow AI across cloud platforms and enterprise applications addresses a governance gap that affects every organization deploying AI at scale. 

The platform’s long foundation in enterprise process logic is a durable competitive asset. Competitors entering ServiceNow’s core territory will find that encoding business process knowledge is slow and expensive work, and that ServiceNow has a significant head start.