For decades, IT service management (ITSM) has followed a familiar model: detect a problem, create a ticket, assign responsibility, investigate, coordinate a response, and document the resolution. Monitoring, observability, asset management, and automation tools support the process, but people connect the information and move work forward. That assumption is now being tested.
As AI evolves from recommending actions to executing them, what happens when technology can identify, investigate, and resolve an issue without someone managing every step? ServiceOps must move beyond ticket management to coordinate operational intelligence, automation, and governance in pursuit of reliable services.
ServiceOps Has Already Established the Foundation
The Enterprise Management Associates (EMA) 2025 research report, Redefining Modern Service Management: ServiceOps, ESM, and the Rise of AI-Powered Services, surveyed 204 IT leaders. Seventy-seven percent reported active efforts to improve service and operations collaboration or an established ServiceOps function. Another 42% identified automated ticket triage and routing, problem-resolution suggestions, and actions among leading ITSM improvement priorities for the next 6 to 18 months.
Organizations initially pursued ServiceOps to coordinate fragmented teams and technologies. AI now offers opportunities to automate that coordination and increasingly execute the work. But fragmented information, integration complexity, and service disruptions remain. As I explored in AI Hasn't Earned That Authority Yet, gaps in operational readiness and access to trustworthy asset data can constrain AI autonomy. Adding AI alone will not solve these problems.
The Ticket Is No Longer the Center of Every Workflow
Consider an infrastructure incident. Monitoring detects abnormal application behavior; observability suggests a cause; ITSM creates and assigns an incident. Engineers investigate, consult configuration information, and execute a change. Each tool contributes value, but people coordinate the response.
Alternatively, an AI-enabled ServiceOps environment could correlate telemetry and recent changes, identify affected services, evaluate an approved remediation, execute it, and verify recovery.
Operational records and controls remain necessary, but the ticket may document decisions, actions, and outcomes rather than coordinate every step. Traditional ITSM metrics emphasize how efficiently organizations process work. Autonomous ServiceOps may eliminate some of that work altogether. A faster ticket resolution is valuable. Preventing a ticket from becoming necessary may be more valuable.
Distributed Execution Changes Service Accountability
This evolution connects with my EMA colleague Dan Twing's EMA Radar for Enterprise Orchestration 2026 blog series. Dan examines the emerging Enterprise Control Plane: how organizations coordinate distributed automation across specialized systems without replacing every execution technology. Visibility, governance, and assurance become essential to understanding enterprise outcomes.
That raises a ServiceOps question: When an AI agent identifies an incident, an automation platform executes remediation, observability verifies technical recovery, and ITSM maintains the record, who is accountable for the outcome?
A technically successful change may fail to restore the business service. A closed ticket may coexist with degraded user experience. ServiceOps must connect execution with service impact, risk, and verified results. That makes service accountability more important, not less.
From Workflow Automation to Governed Operational Action
Traditional workflow automation executes predefined activities. Agentic AI can interpret conditions, evaluate alternatives, and select actions. But the ability to act does not establish the authority to act.
Recent comments from Microsoft CEO Satya Nadella underscore this distinction. In an October 2026 interview on the 20VC podcast, he acknowledged that he no longer believes AI alignment will necessarily improve as models become more capable. He emphasized the need for safeguards around AI systems.
In my earlier blog, Before You Trust AI, Ask These Four Questions, I outlined a framework for evaluating AI adoption and expanding its authority through demonstrated performance. Those principles also apply to enterprise ServiceOps, where autonomous actions can affect critical business services.
Model alignment and enterprise operational governance are distinct challenges. For ServiceOps, increasingly capable agents still need trustworthy context, enforceable policies, appropriate human oversight, and independent verification. An agent might be able to restart a service or initiate a rollback without knowing whether that action is authorized or safe for the affected business service.
Organizations must determine which actions AI may execute independently, which require approval, and how they will document decisions and verify outcomes. Accurate asset and dependency information provides context; observability supplies evidence of operational conditions; automation executes approved actions; and service management connects those actions to policy and business impact.
This raises a critical question for IT leaders: As AI becomes more capable of resolving operational problems, are organizations becoming equally capable of controlling the actions it takes?
The Next Role for ITSM
I don't expect autonomous operations to make ITSM irrelevant. I expect them to challenge the assumption that managing tickets and workflows defines its primary value. As AI handles more investigation, coordination, and execution, ITSM must emphasize authority, service context, risk, and outcomes.
This shift also challenges software categories and vendor business models. Tools that mainly route information or coordinate repetitive tasks may face consolidation pressure. Technologies providing trusted operational intelligence, specialized execution, governance, and outcome verification may gain value, even when users interact with them less directly.
The objective is not to maximize autonomous actions. It is to reduce unnecessary work while improving reliability, accountability, and business value. Autonomy without trustworthy context can accelerate mistakes; execution without verification can appear successful without improving service delivery.
What's Next: Which Tools Still Matter?
As AI-driven execution expands, boundaries among ITSM, ITOM, ITAM, observability, and automation will evolve. Some capabilities will become embedded in broader platforms; others will remain specialized behind common interfaces.
Which ServiceOps technologies will remain essential, which face commoditization, and where will operational software value migrate as AI assumes greater responsibility for execution?
ITAM is one of the first places this shift is showing up. See where organizations stand in EMA’s 2026 maturity benchmark: https://www.enterprisemanagement.com/product/itam-in-the-age-of-ai-emas-2026-maturity-benchmark/
