In our 2025 State of ITSM Report, we showed that generative AI is no longer a science experiment; it is a measurable efficiency engine. Teams using GenAI cut resolution time by 4.87 hours per incident on average, multiplied by the total number of tickets, which could mean hundreds of thousands of reclaimed dollars for a typical mid-sized team. That is not a rounding error; it is a new line item in the IT budget.
But there is a catch: reclaimed time does not automatically turn into business value. Many teams discovered a new kind of efficiency gap in 2025. The gap is the difference between what AI can do on paper and what organizations are structured to leverage. In 2026, that gap is less about tools and more about talent.
Think back to the death by a thousand papercuts that defined legacy ITSM: password resets, access tweaks, approvals, and updates that quietly consumed the workday. In 2025, we proved AI can heal many of those cuts. In 2026, we are realizing you still need a new kind of doctor to run the clinic.
Speed is becoming a commodity. Strategic capacity is the real competitive advantage.
If you want to move from faster tickets to genuine business transformation, you cannot stop at enabling AI for your existing roles. You need to formalize new roles that manage AI itself, the pit crew behind the tools your agents and employees touch every day.
Here are the four new pillars of the AI–first service desk.
Four new pillars of the AI‑first service desk
-
The Knowledge Architect
AI is only as good as the knowledge you feed it. When you ask a virtual agent for help, you are not really testing the model; you are testing the quality, structure, and freshness of your content.
The Knowledge Architect is the person who treats knowledge as a product, not a byproduct.
The profile
This is a detail‑oriented strategist who cares more about data fidelity than data volume. They may come from content management, technical writing, or even library science—any background where taxonomy, version control, and source of truth are the main event, not an afterthought.
Core duties
- Curating machine‑ready knowledge: structuring articles, FAQs, and runbooks so LLMs can parse them reliably, not just people.
- Managing knowledge drift by pruning or updating content as systems, policies, and software versions change.
- Auditing AI responses for hallucinations and misalignment against the official source of truth, then feeding those findings back into content improvements.
Why it matters
Generative AI reflects the data it sees. If your knowledge base is outdated, conflicting, or scattered, your automation becomes a liability instead of an asset. This is how teams end up with AI confidently recommending the wrong VPN client or referencing a retired HR policy.
Without a Knowledge Architect, every AI initiative sits on top of unstable ground.
The ROI
In our State of ITSM data set, the big gains came from more accurate, first‑time resolutions—teams using GenAI are resolving incidents up to 30.5% faster than their non‑AI peers. A dedicated Knowledge Architect helps protect that advantage by reducing incorrect or incomplete resolutions that force agents back into the same tickets, eroding the benefits of those 4.87 hours saved per incident.
When you are reclaiming tens of thousands of hours and hundreds of thousands of dollars in productivity each year, even a modest reduction in do‑over work has a meaningful financial impact.
-
The Conversation Designer
Most AI projects live or die at the interface. If the experience of talking to the bot feels confusing, slow, or tone‑deaf, users will bypass it and go straight back to phones and email, no matter how powerful your underlying models are.
The Conversation Designer is the person who makes interacting with AI feel like getting real help and not wrestling with an unhelpful tool.
The profile
This is part UX designer, part copywriter, part psychologist. They think in intents, not just keywords. They care deeply about how it feels to ask for help and how quickly someone understands they are in the right place.
Core duties
- Mapping intent journeys so the AI understands the difference between “I need a VPN” (a request) and “My VPN is broken” (an incident).
- Designing the voice of your virtual agent to match your culture whether that’s highly professional, more conversational, or something in between.
- A/B testing prompts, flows, and response styles to reduce bot abandonment and increase successful self‑service outcomes.
Why it matters
Self‑service deflection is only real if users choose self‑service. Large‑scale studies show that AI‑driven portals can push self‑service adoption from under 20% to more than 60% when they are designed well, taking real pressure off technicians. The opposite is also true: if users try the bot once and get frustrated, they are unlikely to come back.
Your efficiency gap is not just about what AI can theoretically do; it is about how many people will realistically use it.
The ROI
Our report data show a clear efficiency divide between teams that use GenAI and those that do not. Primarily, GenAI-enabled teams resolve incidents 9.91 hours faster than non-AI teams, a 30.5% relative improvement. The Conversation Designer turns those theoretical savings into actual behavior change by improving adoption and keeping users in the AI channel long enough to see it work.
If you are planning around the tens of thousands of hours reclaimed for a mid‑sized team, a well‑designed conversational experience is what keeps that projection from becoming an over‑optimistic slide in a presentation.
-
The AIOps Orchestrator
A common pattern in early AI projects: the virtual agent can explain how to do something, but a person still has to go do it. That is useful, of course, but it is not the same as end‑to‑end automation.
The AIOps Orchestrator is the role that moves AI from chatting to doing.
The profile
This is a technical generalist who can think across systems. They are comfortable with APIs, JSON, webhooks, and event‑driven architecture. They may come from DevOps, automation engineering, or systems integration, and they act as the glue between SolarWinds Service Desk and the rest of your stack.
Core duties
- Building workflows that connect SolarWinds to tools like Azure AD, Slack, Jira, and beyond so actions can flow directly from AI decisions into real changes.
- Managing trigger logic: deciding when the AI should act autonomously and when a real person needs to be in the loop for approvals or high‑risk changes.
- Monitoring the health of the automation fabric, making sure pipelines and integrations keep working as systems and permissions change.
Why it matters
The difference between “the bot tells you how to reset your password” and “the bot securely resets your password for you” is the difference between guidance and genuine automation. It is also the difference between saving minutes and saving hours.
Without someone owning orchestration, you end up with intelligent suggestions trapped in chat windows instead of flowing into real, measurable outcomes across your environment.
The ROI
In our analysis, the 4.87 hours saved per incident after enabling GenAI come from more than just better answers. They come from removing manual, swivel-chair tasks across apps and systems.
The AIOps Orchestrator is the role that consistently captures those gains, turning one-off automations into a durable automation fabric that can scale with your ticket volume and your business.
-
The AI Governance & Ethics Officer
As AI adoption accelerates, the service desk’s risk profile is changing. It is no longer just the place where tickets live; it is increasingly a place where sensitive data is processed, decisions are made, and actions are taken automatically.
The AI Governance & Ethics Officer is responsible for making sure you move fast without leaving your risk and compliance teams behind.
The profile
This role typically comes from risk management, security, or compliance, with a focus on privacy and data protection. They understand regulatory frameworks, audit expectations, and the realities of day‑to‑day IT operations.
Core duties
- Ensuring AI tools do not expose sensitive HR, financial, or personal data in generalized responses or logs.
- Managing shadow AI by discovering unmanaged tools, bringing them under the secure SolarWinds umbrella, or decommissioning them when necessary.
- Owning AI audit trails, making sure every AI action is logged, explainable, and reviewable for internal and external audits.
Why it matters
In 2026, unmanaged AI may be the single largest emerging risk in many enterprises. External research is already showing a pattern: workers save significant time with AI, but they also spend additional hours reviewing or correcting outputs and worry about data exposure when tools are not centrally governed.
Without a clear governance owner, AI projects can quietly incur hidden costs in the form of security incidents, regulatory exposure, and erosion of employee trust.
The ROI
Data breaches, compliance failures, and reputational hits can erase the financial upside of AI faster than any misconfigured workflow. The Governance & Ethics Officer protects the ROI story by preventing those tail risks from becoming front‑page problems, ensuring the hours saved and dollars reclaimed stay on the right side of the ledger.
Stop Paying the Status Quo Tax
When CIOs look at AI in ITSM, the first question is often: “How many technicians can I cut?” It is the wrong question.
The better question is: “How quickly can I transition my best people into these four roles?”
You have already seen what AI can do for individual incidents: nearly five hours reclaimed per ticket, double‑digit percentage reductions in resolution time, and a growing efficiency gap between AI adopters and everyone else. The next wave of value comes from how you reallocate that time. You can move that time away from repetitive work and into roles that shape, govern, and extend what AI can do across your business.
By moving talent into these modern roles, IT can evolve from a cost center fighting papercuts to a value center orchestrating business flow. The organizations that make that shift first will not just have faster tickets; they will have a structurally different capacity to handle change, complexity, and growth.
If you want to model what this could look like in your environment, don’t stop at reading the headlines. Download the 2025 State of ITSM Report, plug your own ticket volumes into the ROI formulas, and use those numbers to make the case for your AI pit crew.
You may also like
