Atlassian today announced the launch of Atlassian Intelligence, the company’s AI-driven “virtual teammate” that leverages the company’s own models in conjunction with OpenAI’s large language models to create custom teamwork graphs and enable features like AI-generated summaries in Confluence and test plans in Jira Software, or rewriting responses to customers in Jira Service Management.
These new features will only come to Atlassian’s cloud-based offerings. The company doesn’t currently have plans to bring it to its data center editions.
Every company, it seems, is trying to add ChatGPT-enabled features to its service these days, but few companies have the kind of reach and mindshare as Atlassian, especially with developers. Over the course of the last few years, the company also branched out well beyond its original focus on developers to include IT departments and other teams that interface with developers. This now gives it a rather unique view into how teams collaborate, something it is now also leveraging for this new product.
Atlassian notes that the AI system also looks at how teams work together in order to create a “custom teamwork graph showing the types of work being done and the relationship between them.” This data can be enriched with additional content from third-party apps.
For the most part, though, Atlassian Intelligence provides users with a Chat-GPT like chatbox that’s deeply integrated into the different products and that allows users to reference specific documents. For instance, if you want it to summarize the action items from a recent meeting, you only have to tell it to generate a summary and link the document with the transcript in order for it to generate a list of decisions and action items from this meeting — and you can do that right inside of Confluence, for example.
It’ll also happily will draft social media posts about an upcoming product announcement based on the product specs in Confluence.
Similarly, in Jira Software, developers can use the new AI features to quickly draft test plans based on what it knows about a given operating system or other information in a product’s specs.
Users of Jira Service Management, though, may be the most likely to save time with Atlassian Intelligence. Here, users can now use a virtual agent to help automate support interactions right from inside Slack and Microsoft Teams. This new agent will be able to pull up answers from existing knowledge base articles for both agents and end users, for example, and it will also quickly summarize previous interactions for newly assigned agents to bring them up to date on a given issue.
Another nifty feature here is that the new tool can translate natural language queries into the Atlassian’s SQL-like Jira Query Language, opening up this capability to many more users.
All of these new capabilities are now available in early access. Organizations that want to try them can join a waitlist to get access to them here. Following the early access period, some of these features will become paid features over time, but Atlassian specifically notes that the virtual agent for Jira Service Management will be included at no extra cost in its Premium and Enterprise plans.