Agentic ITSM Arrives Before Most Enterprises Are Ready
Vendors like ServiceNow and Ivanti are racing ahead with agentic ITSM tools, even as most enterprises struggle with enterprise AI readiness. Ivanti’s autonomous service desk agent, launched in April, can create incidents, submit requests, and search knowledge bases without human intervention. CIOs adopting these capabilities report shifting teams away from repetitive requests toward higher‑value initiatives, while leaders like Ivanti’s CEO argue the human‑heavy status quo is unsustainable. Agentic AI promises faster ROI and accelerated IT service automation by handling high‑volume, low‑complexity work at scale. McKinsey research highlights one multinational that automated up to 80% of roughly 450,000 annual tickets after moving to agent‑led resolution, redeploying half its service team while maintaining strong customer satisfaction. However, that success followed a deliberate redesign of workflows and customer journeys. The technology worked because the environment was engineered for agents first, not retrofitted after deployment.
The Infrastructure ‘Plumbing Problem’ Blocking Agentic ITSM Tools
Despite the hype around ServiceNow agentic AI and similar offerings, most organisations remain stuck in pilot mode. McKinsey reports that 62% are still piloting AI, with no more than 10% scaling agents in any business function. The core issue is a “plumbing problem”: legacy IT estates are built for ticket‑based, human‑led workflows, not autonomous agents that must cross systems, execute actions via APIs, and operate under strict governance controls. This gap is particularly visible in complex environments such as unified communications, where resolving a single call quality incident might require network telemetry, device health, carrier status, and ITSM data simultaneously. If an agent cannot traverse all these systems reliably, it hands the ticket back to humans—negating the promised efficiency. Leaders like Red Hat’s CEO describe a necessary “back‑to‑basics” phase, where enterprises must relearn fundamentals such as patching and integration before layering on agentic ITSM tools.
Trust, Visibility, and the Observability Gap for Agentic AI
Enterprise AI readiness is not only a technical challenge; it is also about trust and visibility. Traditional monitoring shows whether a service is running, but not what an AI agent decided, which systems it changed, or why. Gartner analysts warn that this lack of transparency makes scaling agentic ITSM tools risky. Unlike deterministic software, AI decision‑making is often opaque, yet its errors can trigger financial loss, reputational damage, or regulatory scrutiny. Gartner forecasts that 40% of organisations deploying AI will have dedicated observability tooling by 2028, leaving a majority without robust monitoring for at least the next two years. In that window, enterprises risk accumulating inference costs and silent failures. Without clear audit trails and explanations for agent actions, operational teams struggle to validate outcomes, and governance bodies hesitate to grant agents wider autonomy—slowing IT service automation and undermining confidence in ServiceNow agentic AI and similar systems.
Rising Cost Pressures Make Agentic ITSM a Necessity, Not a Luxury
Economic pressure is amplifying the urgency around agentic ITSM tools. McKinsey expects IT infrastructure costs to increase two to three times by 2030 as AI workloads grow, even as budgets remain largely flat. In that context, the ability of agents to reduce operating costs and scale IT service automation is becoming a strategic necessity. Yet when deployed into unprepared environments, agents can produce the opposite effect, generating new tickets instead of resolving them. If agents act on stale configuration data, lack access to stable APIs, or operate without clear policy boundaries, errors compound quickly. Inference costs rise without corresponding value, and human teams must step in to clean up mistakes. Enterprises that treat agentic AI as a plug‑and‑play solution risk cost overruns and operational instability. Those that invest first in modernising their IT “plumbing” are better positioned to capture sustainable ROI from ServiceNow agentic AI and similar offerings.
From Pilot to Scale: Prerequisites and Early‑Mover Advantages
Successful early adopters show that the readiness gap is surmountable—but only with disciplined groundwork. McKinsey highlights four prerequisites for effective IT service automation with agents: a sufficiently accurate CMDB for agents to trust configuration data; actions exposed through robust APIs with embedded policy checks; a clear governance model defining what agents can and cannot do; and active monitoring of inference costs and outcomes. For most ITSM teams, these are not simple checklist items but multi‑phase programmes of work that must precede procurement. Organisations that complete this foundation can safely grant agents wider autonomy, rapidly automate high‑volume service requests, and redeploy human talent toward strategic initiatives. As Ivanti and ServiceNow agentic AI offerings mature, early movers who invest in infrastructure, observability, and culture today are likely to gain a durable advantage in IT service automation—while slower peers remain stuck in pilot purgatory, unable to scale beyond experiments.