nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

AI best practices for using Atlassian Intelligence in Confluence 🤖

Sumeeta Kumar

👋🏽 Hi everyone! Sumeeta from the Confluence Product Marketing team.

We know a lot of you are excited about the new AI features and/or have already started playing around with them. We wanted to share some of our tips and best practices that will help smooth the process and optimize your experience.

These include:

How to write effective prompts
How to break down complex requests into smaller, more manageable tasks
How to review and edit AI-generated output
How to work smarter, not harder, with Atlassian Intelligence

Check out our AI guide to learn more. What have been some of your favorite ways to get the most out of Confluence AI? We'd love to hear from you!

Comments

Andy Gladstone

@Sumeeta Kumar thanks for publishing this, I have bookmarked the link and I am looking forward to learning how to better utilize AI in Confluence and on the Atlassian Platform in general.

Miguel ANIA ASENJO

Hello @Sumeeta Kumar

Thanks for the great article.

Regarding this specific question and answer in the AI feature FAQ: https://www.atlassian.com/trust/atlassian-intelligence#does-openai-store-atlassian-customer-data

-> I understand from this statement that OpenAI does not store inputs and outputs submitted for each query? However, we would like to understand how the content of the pages used to provide answers is provided to the OpenAI LLM. I gather that the LLM needs the content of the pages to provide output based on this content. Therefore, is the relevant content sent together with each input, for each individual prompt? If so, how do you determine the relevant content, among all the pages the user providing the input has access to? I cannot imagine every request being sent with all the content, because that would be likely to generate a lot of latency...

Kind regards,

Miguel

Christian Zamora

Is the AI only applicable for internal or if there is an external AI tool for a customer-facing instance?

Mikael Helldén

Hi @Sumeeta Kumar
Any recommendations on how to create content in Confluence that "works well" with AI? In particular, I'm thinking about how to create content that provides excellent definitions when the "Define" feature is used.
Cheers!

Welcome!

It looks like you're new here. Sign in or register to get started.

Quick Links

Find more posts tagged with

confluence

cloud

confluence-cloud

atlassian-intelligence