More Speed, Less Clarity: Why AI Adoption Needs a Visibility Strategy

There’s a version of AI adoption that looks great on paper. Individuals are producing more. Tasks that used to take a day take an hour. Outputs that used to require a specialist can now be generated in seconds. By the numbers, your team looks more productive than ever.

So why does it feel like no one knows what anyone else is actually working on?

This is the question sitting at the centre of Atlassian’s State of Teams 2026 report, and it’s one we think deserves more attention than it’s getting.

The two engines of AI ROI

Atlassian’s research draws a useful distinction between two engines that drive real returns from AI: execution and coordination.

Execution is the bit most organisations have focused on. It’s the stuff that’s easy to see: faster content creation, quicker code reviews, automated summaries, tasks completed in less time. Teams have invested in tools, trained people to use them, and broadly succeeded at this part.

Coordination is the bit that’s falling behind. That’s how well teams communicate intent, share context, align on priorities, and make sure the work they’re doing is connected to something that actually matters.

The report puts a number on what happens when coordination lags: an estimated $161 billion annual cost to the Fortune 500, driven by duplicative work, misaligned priorities, and the kind of coordination chaos that erodes whatever efficiency gains AI delivers. They call it the fragmentation tax.

AI doesn’t create alignment. It amplifies whatever already exists.

Here’s the thing nobody wants to say out loud: AI is great at helping individuals execute. It’s not great at helping teams coordinate. That’s not a failure of the technology. It’s just not what it was designed to do.

When someone uses AI to generate a brief, a plan, or a deliverable faster than before, the output still needs to land in a workflow that people understand. It still needs to connect to the priorities leadership has set. It still needs to be visible to the people downstream who will act on it.

If those things aren’t in place, speed becomes a problem. The report found that:

• 87% of knowledge workers say that with everyone in execution mode, they often lack the time or capacity to actually coordinate
• 84% say they have unclear or conflicting goals and priorities
• 71% say they’ve encountered other teams working on duplicative projects

You can have the most productive individuals in the room and still have a team that’s getting in its own way.

The “workslop” problem

The report introduces a term we think is going to stick: AI workslop. It describes the low-quality, unverified outputs that emerge when AI is used without clear knowledge foundations or standards, and when no one is quite sure who owns the result.

Nearly half of knowledge workers surveyed said their AI outputs aren’t reliably high-quality. Almost the same number said their AI use requires a compromise between speed and quality. And a significant number described the extra work that gets created for everyone when an output isn’t fit for purpose.

This isn’t an AI problem. It’s a visibility and alignment problem. When people don’t have a clear picture of what good looks like, who owns what, or how their work connects to the bigger picture, quality suffers regardless of the tool being used. AI just means it happens faster and at greater volume.

Adoption without alignment is just noise

One of the more striking findings in the report: while 85% of knowledge workers use AI, only 29% have actually embedded it into their daily workflows. Most people still treat it as a personal tool rather than something that’s integrated into how teams work together.

That gap matters. It means organisations are carrying the cost and complexity of AI adoption without getting the coordination benefits.

• 62% of knowledge workers say their technology doesn’t support collaboration well, despite heavy AI investment
• 45% think their leaders view AI as a magic switch that will solve problems without anyone having to change how they work

That’s a setup for disappointment.

Visibility is what makes AI work for teams, not just individuals

The organisations that cut the fragmentation tax in Atlassian’s research did three things differently.

• They gave AI full context to work with
• They designed clear workflows for both people and agents to operate within
• And they built cultures that treat human-AI collaboration as something that needs to be actively shaped, not left to chance

What all three of those have in common is visibility. Clear context requires that people can see the work. Clear workflows require that priorities are connected to execution. A healthy collaboration culture requires that leaders have a line of sight into what their teams are actually doing, not just what they’re producing.

This is exactly the conversation we’re having with organisations across ANZ right now. AI investment is up. But the question of how to ensure that investment delivers at the team level, not just the individual level, is one a lot of leaders are still working through.

Join us on 17 June

On 17 June, we’re hosting a live webinar exploring what it takes to close the gap between strategy and execution: what changes when leaders can actually see the work, and how organisations are connecting team-level activity to the goals that matter.

You’ll hear from Clayton Coetzee, Senior Consultant and Team Lead at Elegance Group, alongside Brandon Huang from Asana and Lachlan Drummond, Head of Value Engineering at Woolworths Group, who will share what this looks like in practice inside one of Australia’s most complex retail organisations.

If the fragmentation tax sounds familiar, this session is worth your time.

Register here