Webinar Q&A - 5 essential tips for AI service management

[#1]

👋 Hello, Atlassian Community! First, thank you so much for your interest and engagement in our recent webinar, 5 essential tips for AI service management at high velocity! We are so appreciative of your time. Don’t forget ❗️ to register for the next 2 webinars in our AI-powered ITSM series here❗️

As a follow up to the first webinar, our team of Jira Service Management, AI, and Rovo experts have taken the time to go through most of your questions and provide answers. Before we get into the questions, here’s a quick summary of the 5 tips from the webinar (and if you missed it, feel free to watch the webinar at the link above, now available on-demand!).

1. Understand the value: AI-powered ITSM has several benefits, including effortless and fast implementation, improved service management, market growth + leadership, enhanced user experiences, and even cost savings.

2. Overcome common challenges: such as data quality and availability, user adoption, integration capabilities, and knowledge management gaps. You can overcome many of these challenges by choosing a solution that unifies data across apps, work, and teams.

3. Focus on governance and compliance: by implementing privacy controls that meet your needs, and choosing a solution that meets compliance standards, makes it easy to audit, and prioritizes transparency.

4. Use AI for ITSM: by taking advantage of features like smart summaries, AI answers that meet your help-seekers where they are, AIOps functionality, and much more.

5. Measure the impact: using metrics like MTTR, time in progress, first contact time, or volume metrics like issues completed per agent.

Without further ado, please see the answers to your questions below!

Q&A

Question	Answer
Licensing
What AI features are available in standard/premium/enterprise?	Jira Service Management is now part of the Atlassian Service Collection, announced at Team Europe. Service Collection offers a variety of AI powered features in the Standard, Premium, and Enterprise editions. From the Virtual Service Agent to Alert Grouping – AI is embedded throughout the platform to assist your teams in providing seamless operations and support. Some features, like Rovo, come with a set credit amount per plan (see more here). You can check out our Jira Service Management AI Feature Guide for an overview of Jira Service Management-specific AI features!
What are the pricing plans and costs? How is Rovo priced?	Jira Service Management is now part of the Atlassian Service Collection, announced at Team Europe. Please see full details here for pricing. Please see full details here for Rovo pricing.
How do Rovo usage limits work?	Rovo comes with a set credit amount per plan, please see more here.
Privacy/Permissions
What data is being used by AI? Which region is it stored in and processed?	Data residency support is available for Rovo. With data residency for Rovo turned on, all of your in-scope app data will remain stored in the region you've selected. To initiate a request to pin in-scope app data for Rovo, review our documentation.
Is my confidential and personal data used by AI protected by GDPR?	We are committed to helping our customers stay compliant with GDPR and their local requirements. As we do today for all of our apps, we will process and transmit data for Rovo in accordance with our Privacy Policy, Data Processing Addendum, and GDPR commitment.
Is your AI hosted by Atlassian or a third party? What did you use for data training if our data is not used? Which LLM does Atlassian use?	We use a diverse range of open-source, Atlassian-hosted LLMs, including models from the LLama series and Mixtral, alongside third-party hosted LLMs from OpenAI's GPT series of models, Anthropic's Claude series of models, and Google's Gemini series of models, to deliver the best outcomes for customers. Our features use dynamic routing to select the appropriate mix of models that can deliver the best experience and accuracy for each scenario. The LLM providers we use do not retain your inputs and outputs, or use them to improve their services. Please refer to our list of data sub-processors for more information on our third-party hosted LLM providers. You can also learn more about how each feature uses LLMs on our transparency page.
Can I give specific teams or team members access to AI and restrict for others? I.e., can it be restricted to product admins? What about Rovo?	Rovo respects all of your existing permissions. In fact, two users may receive different results based on the content they have access to. The data a user has access to is not limited to the app they’re working on. Due to the connected nature of our apps, as long as a user has access to a Jira work item or Confluence page, information can be pulled from across those experiences to inform a response (or output). This policy extends to any third-party connector you have in Rovo. All permissions set in our platform and your third-party connector source will be respected as long as they are set accordingly.
Can I choose to enable only Rovo, or do I have to enable AI in general in order to enable Rovo?	AI is available and automatically activated for all products on Standard, Premium, and Enterprise plans. If you’re not ready for AI, organization admins can deactivate it. Rovo Apps are a core part of the Atlassian Cloud Platform, similar to other apps (like Projects and Goals). These Platform apps are not removable. However, Organization admins can manage (activate or deactivate) AI-powered Rovo features for Atlassian Apps in Atlassian Administration. Please note that non-AI powered Rovo features, such as Rovo Search, cannot be disabled. You can learn more about managing AI in your apps in our documentation. To see a list of Rovo features available with AI activated, see our documentation.
Can I restrict AI to one Atlassian product? I.e., enable it on Jira Service Management but restrict it on Jira? What about Rovo?	Yes, you can manage which apps have AI enabled, including Rovo. More details here.
Can I add company-specific governance rules to Rovo Chat in addition to general permissions?	Rovo permissions are inherited from Atlassian platform settings and any third party connections.
For Rovo and other AI tools available, how much of that is accessible to "clients" within our system (e.g. Confluence)? Can a client use the AI tools too?	Rovo is available for licensed users on a Standard, Premium or Enterprise plan.
Virtual service agent
Is there a resource, framework or a model available on how to best construct the Confluence knowledge base and structure it so that the base virtual service agent in Jira Service Management has the easiest possible job to find correct articles and information?	Yes, check out some tips for setting up your knowledge base to optimize the virtual service agent’s answers here.
What’s the difference between the virtual service agent and Rovo?	Virtual Service Agent A GA capability in Service Collection Only available to Service Collection Premium & Enterprise customers Useful when customers need a higher degree of control over AI’s output (via Intent Flows / very structured approach) - the only way to set intents is through virtual service agent Has its own consumption-based pricing model based on assisted conversations. Customers can execute 1,000 assisted conversations per month or 12,000 assisted conversations per year for free. Above this limit, assisted conversations will start at $0.30(USD)/assisted conversation/month with volume discounts. Rovo Service (Announced at Team EU) Will soon become available to Service Collection Standard, Premium & Enterprise customers It is the umbrella Atlassian-built 1P AI agent designed for service use cases (IT, HR, etc.) Handles incoming requests/tickets, deflects through self-service, resolves complex requests, streamlines workflows like onboarding Same credit-based pricing model as Rovo. Usage limits exist but aren’t currently enforced—billing tools are coming, and customers will get ~90 days’ notice before invoicing begins Note: Virtual Service Agent will be a capability within Rovo Service but it has it’s own pricing model and is only available in Premium+ Rovo Custom Agents Currently in Open Beta and available to Standard, Premium, & Enterprise Jira Service Management customers Customers can create custom AI agents via Studio and embed them in Jira Service Management help channels (Portal, Help Center, Slack, etc.). Leverages Rovo + Teamwork Graph intelligence with: • Simplified configuration • Deep customization (tone, request handling, etc.) Same credit-based pricing model as Rovo. Usage limits exist but aren’t currently enforced—billing tools are coming, and customers will get ~90 days’ notice before invoicing begins
How do I figure out which AI agent type to use?	You can view the details about our out-of-the-box agents here!
Is there a way to make the virtual service agent aware of both scheduled service outages, as well as current unplanned outages/disasters and answer questions about them accurately?	If upcoming outages are published to Confluence, the virtual service agent can reference that data for end users via an Intent or AI answers.
Can the virtual service agent also create a better summary for tickets?	You can use AI on tickets to create a summary. This works great alongside the virtual service agent - for complex intent flows where the virtual service agent asks multiple questions to gather information from the help seeker before opening a ticket, AI summaries can help the assigned agent quickly digest any issue context the virtual service agent has captured. See how to generate an AI summary here.
Is there a way for the virtual service agent AI to create a ticket and self close? Can we use AI to help end users raise tickets to get them allocated to the correct service in the service catalogue?	The virtual service agent can open a ticket and automation can close the ticket based on specified triggers. Yes, Atlassian AI can assist end users in raising tickets by analyzing their requests and automatically categorizing or routing them to the correct service in the service catalogue. This AI-powered ticket triage helps ensure that requests are allocated to the appropriate team or service, improving response times and accuracy.
Can an agent engage with a client through the same virtual service agent chat session?	When the virtual service agent does not resolve the issue, a work item is opened and assigned to an agent who can take over from there.
Rovo
Can Rovo agent help keep up to date articles in Confluence? How well is Rovo integrated with Confluence?	Rovo is well integrated with Confluence and can help keep articles up to date by suggesting improvements, summarizing content, and identifying outdated information using Atlassian Intelligence features. While Rovo can assist with content updates and recommendations, actual editing and approval of changes still require manual review by Confluence users.
Can we track Rovo Chat usage?	Yes! Admins can track usage via the Rovo Insights Dashboard.
Can I use Rovo to update my asset lists?	Rovo currently allows you to search and interact with asset data from Jira Service Management Assets, making it easier to find and reference asset information across your organization. However, direct editing or updating of asset lists through Rovo is not generally available yet—asset management actions like updates or changes should still be performed within the Assets app in Jira Service Management.
When will it be possible to have your custom forms in your intents that Rovo produces and manages?	Atlassian has not yet announced a public release date for supporting custom forms within Rovo-generated and managed intents.
Feature-specific questions & other
What fields are used to generate AI Summaries? I found that Forms are not used. Forms often contain important information. Is it planned to include them into AI Summaries?	Atlassian AI Summaries in Jira Service Management are generated using information from standard issue fields, comments, and attachments, but currently do not include data from Forms. Atlassian is aware that Forms often contain important information, and while there is no public timeline, expanding AI Summaries to include Forms data is being considered for future updates.
Can the AI agent decide what issue types when triaging?	Yes, Atlassian AI can suggest or automatically assign the appropriate issue type when triaging requests in Jira Service Management, based on the content and context of the request. This helps streamline ticket routing and ensures that issues are categorized correctly for faster resolution.
What are the 3 most used Atlassian AI Features in Testing?	AI-powered ticket triage: Automatically categorizes, prioritizes, and routes incoming issues to the right team or agent. AI-generated summaries: Quickly summarizes issue details, comments, and activity to help teams understand context and next steps. AI-driven knowledge suggestions: Recommends relevant articles or documentation to resolve issues faster during ticket handling.
How does AI work cross-functionally between Jira and Jira Service Management?	AI-powered features—such as ticket triage, automated summaries, and knowledge suggestions—can leverage data from both Jira and Jira Service Management, enabling seamless collaboration and faster incident or request resolution across teams.
Could you explain how AI could be used in incident management? What features are there for automated response and tracking?	For incident management, AI can act as an automatic first-level responder using the Confluence knowledge base to answer requests, guide troubleshooting, and even create or update tickets (e.g., change status or self-close). AI assists with ticket triage, categorization, and summarization, helping route incidents efficiently and providing clear overviews for agents. For automated response and tracking, the virtual service agent enables automated chat-based support, ticket creation, status updates, and can answer questions about both scheduled and unplanned outages. AI-driven features include ticket classification, assignment, real-time tracking, stakeholder notifications, and integration with the knowledge base for instant solutions.
Can you communicate with your teams via DevOps?	Atlassian’s DevOps tools (like Jira, Bitbucket, and Opsgenie) enable teams to communicate and collaborate on development, deployments, and incident response through integrated notifications, comments, and automated workflows. While you can coordinate work and share updates within these tools, direct real-time chat is typically handled through integrations with communication platforms like Slack or Microsoft Teams.
Could AI be used for translation and might this be a driver to allow translation for custom field values?	AI-powered translation is possible in Jira Service Management for content such as ticket details or responses, but translation of custom field values is not currently available.
Where can additional information be found about auto-publishing PIRs?	Here is information on creating and publishing PIRs; however, it’s not currently possible to automatically publish PIRs.
Any helpful integrations with Asset Management tools?	Atlassian AI features in Jira Service Management Assets can help by automatically suggesting asset links, surfacing relevant asset information, and assisting with asset-related ticket triage and insights. Additional AI-powered integrations are available through Atlassian Marketplace apps, enabling advanced automation, anomaly detection, and smarter asset management workflows.
Will Atlassian AI be able to read and understand Gliffy flowcharts?	While AI can summarize or answer questions about surrounding text, understanding the structure or logic within Gliffy diagrams is not supported at this time.
Can the chat feature be added as a widget overlaid on my SaaS product?	The Atlassian AI/Rovo chat feature is currently designed for use within Atlassian products and is not available as an embeddable widget for external SaaS products.
Will you be rolling out Arc or Dia browser? What is the timeline?	Atlassian has not yet announced a public timeline for integrating or rolling out browsers with Arc or Dia capabilities following The Browser Company acquisition.
Can we use AI to analyse tickets and make decisions? Some sample of questions: What is the most frequent incident? What is the average time spent to resolve tickets related to a particular component? What is the common solution applied for incidents in a type of component?	Yes, Atlassian AI can help analyze tickets to identify trends such as the most frequent incidents, average resolution times for specific components, and common solutions applied to certain types of incidents. These AI-powered insights enable teams to make data-driven decisions, improve processes, and proactively address recurring issues.
Compared with PIR creation...is there a similar function available when it comes to Problem Management and Known Error Records?	Atlassian AI currently provides automated creation and summarization features for Post-Incident Reviews (PIRs), but there is no equivalent AI-powered function specifically for Problem Management or Known Error Records in Jira Service Management.
Will Jira AI also take into account information in Slack channels when surfacing PIRs and Timelines?	With the appropriate permissions and integrations enabled, Atlassian Intelligence can summarize incident activity and generate timelines using both Jira and Slack data, helping teams capture a complete record of incident response.
Does this replace the need for a "Work Item Collector"?	Atlassian AI and Rovo can help aggregate, search, and summarize work items across multiple Atlassian products and connected tools, reducing the manual effort needed to collect and organize work items. However, if your organization requires highly customized or specialized work item collection and reporting beyond what Atlassian AI and Rovo currently offer, a dedicated "Work Item Collector" tool or process may still be necessary.
If we want to use resources for AI to base answers on that are located outside of the Atlassian ecosystem, is it possible to define specific resources/tools or sitemaps for the AI to get its information from? Is it only able to link internal knowledge bases to JSM or can it also interact with the internet or other AI platforms? Can you link your project to multiple spaces for data and answers?	Atlassian AI and Rovo can connect to multiple internal knowledge bases and Confluence spaces, allowing you to aggregate data and answers from across your Atlassian environment. You can link your project to multiple spaces for broader coverage. While Atlassian AI primarily works with information inside the Atlassian ecosystem, Rovo is expanding its integrations to include select third-party SaaS tools and platforms. However, it does not currently support crawling arbitrary external websites or sitemaps, nor does it interact directly with the open internet for sourcing answers.
Can knowledge base connect to Sharepoint as a repository?	Yes, you can connect Sharepoint to your knowledge base.
Let us suppose we have a knowledge base in Confluence. Can Atlassian's AI act as an automatic first-level response for requests based on the Confluence document? And if so, would it be able to react to the ticket/issue (e.g. change status)?	Yes, Atlassian’s AI-powered virtual agents (such as Rovo or virtual service agent in Jira Service Management) can act as an automatic first-level responder by using your Confluence knowledge base to answer common questions and resolve requests. While the AI can suggest solutions and answer tickets based on documentation, its ability to take actions like changing ticket status depends on your configuration and permissions—automated status changes are possible when set up with the right workflows and automation rules.
Can we use Atlassian AI to clean up existing documentation on confluence? E.g., we have multiple spaces about similar topics and we would like to merge then in one or remove the duplications there.	Atlassian AI in Confluence can help identify similar or duplicate pages and summarize content, making it easier to find and prioritize the most relevant documentation. However, merging spaces or automatically removing duplicates still requires manual review and action—AI can assist with suggestions and summaries, but full automation of documentation cleanup is not currently available.
Is adding tickets to knowledge base articles possible so that it shows possible solutions along with similar tickets? And is the similarity based on summary?	Atlassian’s AI can suggest relevant knowledge base articles and display similar tickets when users create or view issues, helping surface possible solutions based on past incidents and documentation. Similarity is determined using AI analysis of the ticket summary, description, and other key fields—not just the summary—so recommendations are contextually relevant.

More questions? Leave us a comment below and we’ll do our best to answer them!

[Josh]

Hi @amandagit . Thank you for the helpful article.

The terminology around VSAs and RSAs is pretty confusing; I'm seeing it in Community posts and even within my team of org admins. It also seems like VSAs might be sunset at some point in the near future since they offer far fewer capabilities (unusable for my org).

Any chance there could be more terminology and visual differentiation between those two feature sets? For example, VSAs = "Intent Flows" and keep "Rovo Agents" as is?