Introduction

IT service management has always evolved with its tools. From ticketing systems to workflow automation, every generation of ITSM tooling has reshaped how teams deliver value. In 2025, many platforms now incorporate generative AI features, promising faster triage, quicker documentation and more informed change planning. These capabilities, however, are not “magic.” They represent another category of tooling that must be assessed, governed and improved in alignment with ITIL.

As an ITIL Ambassador, I see this discussion as especially important. PeopleCert emphasizes not merely adopting tools, but ensuring they align with practices, guiding principles and governance. Treating AI capabilities as tools—rather than solutions in themselves – allows organizations to evaluate them with the same discipline applied to any ITSM enhancement.

How AI capabilities appear in ITSM tools

AI-enabled ITSM platforms combine multiple types of AI capabilities. Some features rely on established machine learning techniques, such as ticket classification engines that suggest categories and likely owners to reduce manual triage.

Generative AI capabilities build on top of these foundations. They appear in areas such as drafting knowledge articles from ticket histories, generating first-pass change risk notes and checklists, summarizing incidents into plain-language updates for stakeholders, and producing draft runbooks based on past resolutions.

These capabilities are valuable, but they do not replace accountability. Service desk analysts, problem managers and change authorities remain responsible for decisions and outcomes. In practice, AI behaves as an accelerator – saving minutes in classification, hours in documentation or days in knowledge capture – while still requiring human validation.

In one service desk pilot I observed, AI-assisted ticket routing initially led to misrouted urgent incidents because the team trusted the tool too early. Once controls were introduced – requiring human confirmation until accuracy proved reliable – the feature began to measurably improve MTTR. This balance between speed and oversight directly reflects ITIL’s “optimize and automate” principle.

Benefits and trade-offs

The benefits of AI-enabled ITSM tools are clear. They reduce manual effort, improve consistency in records and help smaller teams scale without increasing headcount. Documentation often sees the greatest gains, as AI can convert unstructured notes into structured drafts that teams otherwise lack time to produce.

The trade-offs, however, are equally important. Poor data leads to poor recommendations. Some models generate plausible but inaccurate outputs, which can be more dangerous than providing no suggestion at all. Compliance and privacy risks emerge if sensitive information is processed without safeguards. Transparency is another concern: when tools cannot explain why a decision was suggested, auditability becomes difficult.

These challenges are not reasons to avoid AI-enabled tools. They are reminders that governance and validation must be embedded from the start.

A practical evaluation framework

When assessing AI-enabled ITSM features, organizations should apply the same rigor used for any tool selection:

• Functionality: Which ITSM use cases are supported?
• Accuracy: How well does the tool perform on real operational data?
• Controls: Are human approval, logging and overrides available?
• Data handling: Where is data processed and stored?
• Integration: Does the tool align with existing workflows?
• Measurability: Can improvements be tracked through metrics such as MTTR or FCR?

This structured approach supports governance and continual improvement.

Governing AI-enabled ITSM tools

Governance determines whether AI-enabled tools succeed or fail. Human-in-the-loop should be the default posture. Any AI suggestion that affects services or configurations should require confirmation until accuracy and auditability are proven. Audit trails should capture prompts, outputs and approvals.

Confidence thresholds can guide phased autonomy. For example, high-accuracy triage engines may auto-route low-priority tickets while keeping critical incidents under manual review. Regular validation ensures performance does not degrade over time. Clear ownership is essential—someone must be accountable for both tool performance and data stewardship.

How to adopt safely

A phased adoption approach works best. Start with a narrow use case, such as AI-assisted triage in a single queue. Establish baseline metrics, observe results over several weeks and define clear pass/fail criteria before scaling.

User and staff training is critical. Teams must understand that AI outputs are drafts, not authoritative decisions. When employees are encouraged to question, edit or reject suggestions, they remain active participants in quality control rather than passive recipients of automation.

Closing thoughts

Generative AI is entering ITSM not as a replacement for professionals, but as a new category of tooling. When governed with the same rigor ITIL expects of any tool, these capabilities can enhance efficiency and scalability without compromising accountability.

For ITIL practitioners, the message is straightforward: evaluate carefully, adopt incrementally and measure impact. Done correctly, AI-enabled tooling can strengthen—not dilute—value delivery.

The new PeopleCert report, AI in ITSM tools, explores how AI capabilities are transforming ITSM tools today and what lies ahead for the industry.

For more insights and expert guidance, explore ITIL-aligned ITSM tools and PeopleCert ITIL certifications.