AI governance requires more than policies and committees. Discover why organizations need visibility of their AI agents, clear ownership, lifecycle information and reliable change records to govern AI effectively.
Ticket handoffs between IT, customer support, and engineering can lead to lost context, unclear ownership, and unresolved customer issues. Discover where handoffs break down and how to automate them safely.
Our industry has talked about the data trapped in ITSM tools for years, but the real issue isn’t collection; it’s collecting without a purpose. This piece looks at unused and underutilized ITSM data, why CMDBs are the classic example, and how AI is finally positioned to spot what’s wasted, what’s underused, and what should be collected next.
Many IT service desks are buried in password resets and status requests, and the usual response is a full ITSM platform migration. Aditya Santhanam argues for a cheaper route: wrap conversational AI around the ITSM tool you already run, so AI agents work inside existing records, SLAs, and rules. The article covers seven use cases, a six-step rollout, what to keep humans on, and the metrics that show it works.
An employee reports a dropped VPN connection, and every step of the IT service desk automation fires perfectly: intent read, ticket routed, history attached. The VPN still does not work. This gap between a well-handled ticket and a solved issue is the most common outcome in IT service desk automation today. The article maps a five-level outcome ladder and how to measure verified resolution instead of ticket volume.
AI costs are spiraling across enterprises, and budgets built for predictable per-seat SaaS costs were never designed for open-ended, per-token inference. Erhan Giral of BMC Helix walks through why unmonitored AI agent loops and repeated context re-sends are driving the overrun, and lays out how intelligent model routing can cut inference costs while keeping most of the quality frontier models deliver.
SolarWinds surveyed 844 IT professionals for its 2026 State of ITSM report, and the results cut two ways. Eighty-four percent said AI met or exceeded ROI expectations, saving close to nine hours a week on detection, support, and triage. Yet 52% said overall workload increased, driven by new work maintaining integrations, validating outputs, and training staff. Outcome-based measurement separates the two groups.
Most Slack-based support isn’t a service desk; it’s a search bar and good intentions, and every untracked request is a governance issue waiting to surface. Taylor Halliday walks through building a real ticketing workflow with two channels, Slack’s Workflow Builder, and an emoji status board, then lays out exactly which pains mean you’ve outgrown the hack and need a dedicated tool.
Agentic AI systems build responses nobody specified in advance, and traditional governance models weren’t built for that. Jon Stevens-Hall draws on complexity thinking, from Cynefin to LLM-as-judge testing, to ask what it takes to govern behavior that develops rather than behavior that’s designed, and why loosening control in the right places might be the only way to keep it.
Salil Kulkarni, CSIO at Virima, argues that when an AI agent gets overridden, the issue is rarely the model. It’s a data layer that’s stale, incomplete, or missing the context a human would have caught. Drawing on decades leading technology at Caesars, Las Vegas Sands, and in hospitality revenue management, he sets out four conditions a data layer must meet before an AI agent can be trusted to act.
ITSM AI agent pilots that dazzle in the demo but stall in production are usually blamed on the AI model. The article’s author, a ten-year ServiceNow veteran, argues that the real issue is the CMDB beneath it: incomplete, duplicated, and poorly governed data that any AI agent inherits as its own weakness. Before offering four steps to close what he calls a skills gap, not a technical debt problem.
AI may automate incidents, requests, and knowledge management, but it also becomes a service that requires governance, accountability, and human ownership. Discover why service management remains essential in an AI-driven future and why trust depends on more than intelligent technology.
James Finister has spent a career in service management and now works on AI ethics, and his argument is that the industry still doesn’t value people enough. In episode 11 of Conversations with Giants, he makes the case for governing AI honestly and early, designing work around the people you have, and remembering that SIAM is about relationships and trust, not just the contract.
AI can only be as effective as the service management environment that supports it. This article explains why service management architecture, rather than tools alone, provides the operational foundation needed for AI, Agentic AI, and intelligent enterprise transformation.
For 30 years, the ticket ran the IT service desk: submit, log, triage, resolve, close. Agentic AI in ITSM is retiring that model, executing fixes across systems rather than drafting them for a human to click through. There are five places IT leaders get this transition wrong, from inflated agentic claims to the governance gap most vendor decks skip entirely, and why the ticket itself may not survive the shift.
