Teams | Collaboration | Customer Service | Project Management

AI can sound confident about almost anything. The challenge is making sure it’s right. At Atlassian, Content Technologists are building the systems, standards, and structures that help AI deliver better outputs, removing the burden from our customers of knowing what information to trust.

AI is changing both how employees work and how IT teams manage the tools they rely on. We are building for both. This week, we added two new AI capabilities to CurrentWare Cloud. AI Usage Insights is the workplace side that shows you which AI tools your employees are actually using, how adoption is trending across teams, and where unapproved tools are showing up.

When Andrew King, who heads up competitive intelligence at Asana, needs to help a sales representative win a deal, the information has to be exactly right. This means sourcing specific, current details customized for each individual rep. King supports a 400-person sales team. Each rep is working different deals, facing different competitors, and at different stages of the sales cycle.

Open the Log Time modal in Teamwork.com, click over to the Log my day tab, and the entries are already there: an hour on a client kickoff call, thirty minutes on competitor research, two hours drafting a homepage design. Teamwork.com built each of these from the tasks and projects you actually worked in today, before you typed a single word.

Earlier this year, we wrote about turning Atlassian Design System guidance into structured content that AI agents can actually use. Since then, agents have reached that content through our MCP server tools and an agent skill. We recently added a third distribution method: a command-line tool. The first question from our own engineers was a fair one: Why ship a CLI when you already have an MCP server?

As AI accelerates work, leaders need to be more deliberate about what technology takes off people’s minds and where human attention can create the most value. AI promises to help organisations move faster. But at two Asana executive dinners held in Sydney and Melbourne in August, a different leadership question emerged: what should we choose not to accelerate? The challenge is not finding more hours in the day, but creating better conditions for the hours people already spend.

Most vendors sell "AI work management" as a prettier board with a chat box. That framing is too small for anyone running paid client delivery. AI work management is the operating layer that decides who does what, when you can safely promise dates, and what the work actually costs. It also decides which parts a supervised AI agent can take without wrecking trust. Before I joined Teamwork.com, I saw the spreadsheet version of that job repeatedly across agency environments.