Teams | Collaboration | Customer Service | Project Management

When a tool can do in seconds what used to take hours, it’s natural to wonder if there will be any work left for humans. But new research from Atlassian’s Teamwork Lab suggests the opposite: Rather than shrinking an employee’s remit, AI appears to be expanding both the breadth of what employees take on and the depth at which they operate.

There's a question that consulting leaders don't ask themselves enough: if AI created 10% more delivery capacity in your teams tomorrow, would you actually know where to deploy it? That provocation sat at the heart of the session I ran at Leaders in Consultancy Munich, and judging by the room, it landed.

Most enterprise AI programs don’t fail at the model level. They fail at the foundation — the data strategy, governance model, access controls, and measurement discipline that determine whether a pilot ever becomes a production system. The organizations scaling AI well did the unglamorous work first.

Customers have learned to expect that service should work like a conversation: describe the problem in plain language, get it resolved. Tools like ChatGPT and Gemini set that bar, and most contact centers fall short of it. The obvious fix is to add AI, yet MIT's NANDA initiative found that around 95 percent of enterprise generative AI pilots never reach measurable value. So the question worth asking is not whether to put AI into customer service.

On June 23, 2026, Anthropic launched a product that should keep Slack’s leadership awake at night, and it launched it inside Slack itself. Claude Tag lets any employee mention @Claude in a Slack channel and hand it real work: watch threads, draft documents, fix code, chase forgotten follow-ups. It doesn’t behave like a bot. It behaves like a colleague, one that never sleeps, never forgets what was said in the channel last Tuesday, and never needs to be trained twice.

AI makes your organization’s knowledge easier to find and use. That’s the whole point. But it also means sensitive data moves faster, surfaces in more places, and becomes harder to track. The pressure to act is real, but the playbook isn’t new. You still need to know where sensitive data lives, govern how it’s being used, and prevent it from unauthorized exposure. AI is now giving you a reason to revisit your data security posture and make sure it’s strong enough to keep up.

Less than six months ago, the Atlassian Rovo Model Context Protocol (MCP) server went GA, giving Claude, Cursor, and every major AI agent direct access to Atlassian for our customers. Today, over one million users trust it every month to do real work through agents. But that number isn’t the story. The story is what’s happening inside those interactions: how AI agents are actually being used at enterprise scale, and who’s getting the most value.