AI boosts productivity but fails to transform team outcomes

New research from Atlassian has shown that while the use of artificial intelligence in the workplace is rapidly increasing, many organisations are not seeing significant improvements in team performance or strategic outcomes.

The Atlassian Teamwork Lab surveyed 12,000 knowledge workers across six markets, including Australia, the United Kingdom, the United States, India, France and Germany. The research, which also involved 180 executives at VP level or higher from Fortune 1000 companies, is published as the Atlassian AI Collaboration Index 2025 and investigates how workers and organisations are engaging with AI tools on a daily basis.

Rising adoption

According to the findings, daily use of generative AI at work by Australians has increased more than threefold since 2024. More than half (55%) of Australian respondents now use AI every day for work, with 90% using it at least once a week. This points to widespread adoption of AI technologies among Australian knowledge workers.

Globally, workers say that AI is making them 33% more productive on average, resulting in an average saving of 1.3 hours per day. Australian employees report similar results, saving an average of 78 minutes per day through AI use. However, a key challenge remains: 80% of Australian employees feel held back because their AI tools are not connected to the right data and information, indicating a common bottleneck in AI implementation.

Changing work habits

The way employees seek information is also evolving. The research revealed that 51% of workers are more likely to search for answers using AI rather than asking a colleague (44%). This shift in habits underscores AI's growing role as a primary information source in the workplace, but it may also impact team collaboration and knowledge sharing.

Yet, despite the rapid uptake, transformational results are rare. Globally, only 3% of executives report that AI has led to transformative change within their organisations. Most companies have not experienced dramatic improvements in efficiency, innovation or quality of work. In fact, almost half of the executives surveyed said they had seen only very slight improvements at best, with just 4% confident that AI is helping to solve previously unsolvable or highly complex problems.

Personal productivity vs collaboration

The Index highlighted a paradox: while AI is effective at boosting individual productivity, an organisational focus on personal gains could be costly in the long run. The study indicates that companies concentrating on personal productivity instead of AI-driven coordination are 16% less likely to see increases in innovation. This misalignment could have substantial financial implications, with the report estimating that the Fortune 500 could lose USD $98 billion each year in returns on AI investments if the focus remains solely on personal efficiency.

In the words of Molly Sands, Head of Teamwork Lab at Atlassian:

"AI is everywhere - tackling to-dos, summarising docs, and crunching data, but teams are still drowning in tasks, and now an ever-growing wave of AI tools. Silos persist, work remains scattered across platforms, decisions are fractured, and goals are disconnected. When AI is used to speed up output without fixing the roots of this disconnect, it just amplifies the chaos."

Most knowledge workers acknowledge that senior leadership is increasingly supportive of AI. The proportion of workers who feel their leaders foster a safe environment for AI experimentation rose from 60% to 74% in the past year. Furthermore, 65% of respondents have seen their manager use AI directly in problem-solving or task completion, and workers whose managers model AI use are significantly more likely to use AI throughout the day and collaborate strategically with it.

Developing new skills

The research also sheds light on skills gaps and development priorities. In Australia, 48% of respondents identified prompt engineering - specifically, 'writing clear AI prompts' - as the top AI-specific skill to develop in the next year. Creativity was cited by 45% of Australian workers as the most important general ability to build, reflecting a growing interest in using AI as a tool for innovation.

The time saved through AI is mostly reinvested in areas that benefit both work and personal life. In Australia, workers most commonly used the extra time for 'work-life balance' (49%), 'strategic thinking and planning' (43%), and 'process improvement' (41%).

Despite the potential, only one in three employees globally fully trust AI, and 79% say they would use AI more if it had access to the right data and information. The hesitation underscores the ongoing need to address integration and trust issues associated with these new technologies.

Strategic AI collaboration

The concept of the "Strategic AI Collaborator" emerged as a key theme in the research. In Australia, 43% of respondents were identified as taking a proactive, integrated approach to using AI in the workplace. These workers are more likely to benefit from management's modelling of AI use and are also more likely to leverage AI for broader collaborative efforts rather than solely for personal productivity.

With AI usage increasing sharply and tangible productivity gains evident, the findings suggest that organisations may need to rethink their approach to AI deployment if they are to realise more strategic, system-wide benefits.