While 85% of knowledge workers report using AI tools, only 29% have integrated AI into their daily workflows. Atlassian’s experience shows the key challenge lies in transforming organizational behavior rather than technology itself.
- 85% of workers use AI but only 29% embed it into workflows
- Shifting AI enablement under people leaders improved adoption
- Successful adoption requires top-down and bottom-up change
What happened
Many enterprises struggle to realize AI’s full value because they focus on the technology’s capabilities or integration challenges rather than the organizational changes required. Atlassian’s research and internal experience showed that simply providing AI tools or licenses does not change how employees make decisions or coordinate.
Early AI deployments at Atlassian revealed that while individuals sped up tasks, teams remained fractured and adoption plateaued. Efforts initially concentrated on technology rollout and user metrics like login counts, which failed to translate into meaningful behavioral shifts or team-wide adoption.
Why it matters
The core issue in most stalled AI rollouts is behavioral and cultural resistance, not immature technology. Without embedding AI into daily workflows and transforming how teams collaborate and trust outputs, enterprises cannot unlock AI’s potential.
Atlassian’s breakthrough came from framing AI adoption as a change management challenge and shifting responsibility from IT to the people function. This approach empowered leadership to communicate purpose, grant permission for experimentation, and create safe spaces for trial and error, which proved critical for adoption.
What to watch next
Effective AI adoption requires a dual approach: executive-driven vision coupled with grassroots innovation. Organizations should actively build ecosystems where users can share practical AI use cases and continuously adapt workflows, rather than imposing top-down mandates alone.
Understanding and evolving the organizational culture before AI rollout is essential. If the environment does not currently support experimentation or transparency about AI use, these cultural barriers must be addressed first to pave the way for successful technology integration.