For years, Enterprise Service Management (ESM) has focused on extending IT service management principles beyond IT.
HR requests, finance approvals, legal inquiries, procurement tasks, facilities issues — all can be brought into structured service workflows, with defined request types, ownership, SLAs, approvals, and reporting.
The model works. But it also creates a familiar problem.
Employees still have to know where to go, what to ask for, which form to use, and what information to provide. Service teams, meanwhile, can spend significant time interpreting requests, gathering missing information, routing work, and answering questions that have already been documented somewhere in the organization.
AI is beginning to change that model.
With capabilities such as Jira Service Management's Virtual Service Agent and Atlassian Rovo, ESM is moving beyond simply managing requests. The service platform can increasingly understand what an employee needs, find relevant context, guide the interaction, and help move the underlying work forward.
The result is a shift from service request management toward intelligent service delivery.
From Tickets to Conversations
Traditional service portals put the responsibility on the employee.
Find the right portal. Choose the right request type. Complete the form. Wait for a response.
That process creates structure, but it also assumes that the person requesting a service already understands how the organization is structured.
AI introduces a different interaction model: employees can start with the problem rather than the process.
Jira Service Management's Virtual Service Agent can recognize questions and requests, provide answers from a connected knowledge base, guide users through conversational flows, gather information, and route requests to human agents when necessary.
That distinction matters.
An employee doesn't necessarily need to know whether a request belongs to HR, IT, Finance, or another service team. They can describe what they need, while the service experience handles more of the underlying routing and qualification.
This is one of the fundamental changes AI brings to ESM:
The user no longer needs to understand the service architecture in order to navigate it.
From Manual Routing to Intelligent Orchestration
The next step goes beyond answering questions.
A service request often requires several actions before it can reach the right person. Information needs to be collected, the request needs to be categorized, priorities may need to be established, and the appropriate workflow needs to be triggered.
AI can help automate parts of that process.
Jira Service Management's Virtual Service Agent can use intents and conversation flows to recognize specific requests, gather information and perform defined actions or create work for human agents when intervention is required.
This creates a more useful role for AI than simply acting as a chatbot.
Instead of:
Employee → Form → Ticket → Agent → Resolution
the experience can increasingly become:
Employee → Conversation → Understanding → Action → Resolution
The difference is not just fewer clicks.
It changes where the work happens.
Routine information gathering and qualification can happen before a request reaches a specialist. Human agents can receive a more complete and structured request rather than spending the first part of the interaction figuring out what the employee actually needs.
For organizations operating ESM at scale, this can reduce administrative friction without removing human oversight from processes that genuinely require it.
From Knowledge Bases to Enterprise Context
Self-service has always depended on knowledge.
The problem is that enterprise knowledge rarely lives in one place.
Policies may be stored in Confluence. Project information may sit in Jira. Communication may happen in Slack or Microsoft Teams. Other business systems contain additional operational data.
Even a well-maintained knowledge base can therefore leave employees searching across different systems to find the context they need.
This is where Rovo changes the conversation.
Rovo combines enterprise search, chat and configurable AI agents, using organizational context to help teams find information and complete work. Its agents can work with Atlassian data as well as connected third-party applications, depending on their configuration and available permissions.
The important point is not that AI magically creates a "single source of truth."
It doesn't.
Instead, AI can create a single access point to information that remains distributed across the organization.
That is a much more useful way to think about AI in ESM.
The underlying systems can remain specialized, while the employee experience becomes increasingly unified.
What This Means for ESM Teams
The value of AI-enhanced ESM isn't simply that employees receive answers faster.
The larger opportunity is to change how service teams spend their time.
Less repetitive work
Routine questions, information requests and predictable service interactions can increasingly be handled through self-service and conversational experiences.
Better-quality requests
When AI gathers information before escalation, agents can receive requests with more of the relevant context already captured.
Faster resolution
Agents can spend less time reconstructing ticket history, searching for information or asking basic follow-up questions — and more time solving the underlying problem.
A more consistent employee experience
When different departments provide services through a common service-management model, employees don't have to learn an entirely different process for every request.
Greater scalability
As organizations grow, simply adding more agents to handle increasing request volumes is rarely the most sustainable answer. Automation and AI can absorb part of the repetitive workload while human teams remain focused on exceptions, complex cases and higher-value work.
The result is not an "AI help desk."
It is a service-management model where human expertise and AI capabilities are combined within the same operational framework.
The Governance Question Doesn't Go Away
There is, however, an important caveat.
AI does not make poor enterprise information better.
If policies are outdated, knowledge articles contradict each other, permissions are poorly managed, or processes are unclear, introducing AI can make those problems more visible — and potentially more difficult to manage at scale.
Atlassian's own documentation highlights the importance of a well-structured, up-to-date knowledge base for AI answers in Jira Service Management. AI answers use linked knowledge base content to generate responses, while respecting the relevant access permissions.
This makes knowledge management part of the AI strategy rather than a separate documentation exercise.
Organizations looking to introduce AI into ESM should therefore consider:
- Knowledge quality: Is the information accurate, current and clearly structured?
- Permissions: Can AI access only the information the relevant user is authorized to see?
- Process ownership: Who is responsible for maintaining policies, workflows and service definitions?
- Human oversight: Which requests should always be escalated to a specialist?
- Continuous improvement: Are AI interactions being reviewed to identify gaps in knowledge and service processes?
The organizations that benefit most from AI will not necessarily be those that deploy the most agents.
They will be the ones with the clearest processes and the strongest information foundations for those agents to work with.
Where Should Organizations Start?
AI doesn't require an organization to redesign its entire ESM strategy overnight.
A more practical approach is to start with the areas where the combination of volume, predictability and available knowledge creates a clear opportunity.
1. Identify repetitive requests
Look for high-volume service requests that follow predictable patterns or can be resolved through existing information.
These are often the best candidates for AI-powered self-service.
2. Review your knowledge foundation
Audit the Confluence spaces and other knowledge sources that support your service processes.
Remove outdated content, resolve contradictions, clarify ownership and make important information easier to find.
3. Map the human handoff
Not every request should be automated.
Define where AI can answer, where it can gather information or initiate an action, and where a human specialist should take over.
4. Expand from individual use cases
Once a successful use case is established, look beyond IT.
The same principles can apply to HR requests, facilities, finance, procurement, legal services and other internal functions.
That is where ESM becomes particularly powerful: the service model becomes reusable across the organization rather than remaining confined to the IT help desk.
ESM Is Becoming More Intelligent — Not Less Human
The future of ESM isn't about replacing service teams with AI.
It is about removing the friction between an employee's request and the work required to resolve it.
Jira Service Management provides the structure: service portals, workflows, requests, SLAs, automation and controlled processes. AI capabilities such as the Virtual Service Agent and Rovo add a more conversational and context-aware layer on top of that foundation.
That combination opens a different way of thinking about enterprise service management.
The goal is no longer simply to create a better ticketing system.
It is to create a service environment where employees can describe what they need, AI can help understand and move the request forward, and human teams can focus their expertise where it delivers the most value.
That is where ESM moves beyond the help desk — from managing requests to intelligently orchestrating work.
Ready to Explore AI-Enhanced Service Management?
If you're evaluating how Jira Service Management, Rovo and AI-powered workflows could fit into your organization's service-management strategy, we can help you assess your current processes, identify practical automation opportunities and design an approach that balances efficiency with governance.