Atlassian Corp. Plc today became the latest technology company to add generative artificial intelligence smarts to its flagship collaborative software offerings.

The new technology, Atlassian Intelligence, is based partly on in-house models the company gained via its acquisition of Percept.AI in January 2022. It also leverages OpenAI LP’s GPT-4 model, which notably powers the ChatGPT chatbot that sparked a virtual AI arms race among big tech firms following its launch late last year.

Dozens of software companies have been looking to capitalize on the buzz around generative AI, which enables machines to interact with humans and respond in an almost lifelike way, replying to questions, finding information, performing tasks and more.

Atlassian Intelligence was built using large language models that sit at the heart of generative AI, and works by constructing “teamwork graphs” that show the different kinds of work being done by teams, as well as the relationships between them. The open nature of Atlassian’s platforms further enables it to provide additional context from third-party applications, the company said.

Atlassian said GPT-4, which has been trained on enormous volumes of publicly available text online, will be able to assist teams in multiple ways, the company said, by accelerating work, providing instant help and building a shared understanding of projects.

In the Confluence collaboration platform, workers will be able to click on any term within documents they don’t understand, and the assistant will automatically generate an explanation and provide links to other relevant documents. Users can also type questions into a chat box and receive automated answers derived from the information within documents uploaded to Confluence. Tell it to generate a summary of a recent meeting and add a link with the transcript, and it will instantly spit out a list of decisions and action items that were agreed upon.

In addition, Atlassian Intelligence can draft social media posts about an upcoming product announcement, based on the product’s specifications in Confluence. Meanwhile, software developers using Jira can use the tools to rapidly draft a test plan based on what it knows about a specific operating system.

Jira users can also make use of a virtual agent that automates support through Slack and Teams. The agent would be able to pull information from existing knowledge base articles to help both agents and end users, while summarizing previous interactions for newly assigned support agents, so they’re immediately brought up to date on an issue.

Yet another cool feature is Atlassian Intelligence’s ability to translate natural language queries into Atlassian’s Jira Query Language, which should make it very useful for developers.

“AI is coming to software development — that is simply unavoidable, and Atlassian doesn’t want to be left behind,” said Holger Mueller of Constellation Research Inc. “It’s a good move, infusing ChatGPT capabilities into Confluence and Jira, because anything that increases the velocity of software developers will be welcomed by enterprises. What will be interesting to see is which of the new features become the most popular and useful.”

Atlassian said customers will have to sign up to a waiting list if they want to access the new features, which are now available in beta test mode in its cloud-based products only. Over time, many of the new features will eventually become paid features, though some, such as the virtual agent for Jira Service Management, will be made available at no added cost in Atlassian’s Premium and Enterprise plans.

New users who sign up for the beta can expect to see them start appearing in the next few months, Atlassian said.

Images: Atlassian