Australia Leads Global Peers in AI Agent Use as ‘Agent Sprawl’ Piles More Work onto Managers

• • Nearly six-in-ten (58%) surveyed Australian directors and senior managers expect to manage more AI agents than people within the next 12 months.
  • 59% spend more time reviewing and guiding their team’s output since AI use increased, the highest result across the seven surveyed markets
  • A third (33%) say AI adds hours to their working day, while 62% say identifying and testing where AI creates value currently requires more effort than it saves.

Australian businesses are adding AI agents faster than they are redesigning the work around them, creating a new form of workplace bloat monday.com calls ‘agent sprawl’ - the accumulation of agents and automated hand-offs without clear ownership, controls or measures of value. 

A global study commissioned by AI Workspace monday.com and conducted by Nielsen found 46% of Australian directors and senior managers use AI agents to complete multi-step work, the highest proportion across seven markets, and well above the 38% global average. Nearly six-in-ten (58%) expect to manage more agents than people within the next 12 months. 

But that lead isn’t yet translating into lighter workloads. Nearly six-in-ten (59%) Australian leaders spend more time reviewing and guiding their team’s output since AI use increased, the highest result in the study, while one in three (33%) say their working day has grown longer.

The agents finish the task, but managers inherit the work around it
An agent may complete an individual task in seconds, while the preparation and consequences remain human. Managers must decide what to delegate, provide context, connect the agent to company systems, check the result, correct mistakes and carry responsibility for the final outcome. That surrounding work is spreading across Australian organisations: 

• • 65% of surveyed leaders report using more software tools since adopting AI, the highest share across the seven markets 
  • 58% say their tools have become more interdependent, while 36% have gained additional manual steps in their workflows
  • 36% report an increase in the number of manual steps within their workflows
  • 62% say identifying and testing where AI creates value currently requires more effort than it saves

Experience hasn’t removed the burden - more than two-thirds (67%) of respondents describe their teams as advanced or expert AI users, the highest proportion among global peers, while 69% believe their organisation provides sufficient AI training. The friction appears to sit in how work is organised around the technology, rather than whether employees know how to use it.

“Agent count is fast becoming a vanity metric. Giving one employee ten agents doesn’t make them ten times more productive if they inherit ten streams of work to supervise. Businesses need to measure whether the entire workflow moves faster, with less rework and fewer hand-offs. Every agent also needs a job description, a defined outcome, its own permissions, a clear point of human escalation and one accountable owner. Without that discipline, agent adoption simply creates more work the organisation can’t see,” said Gavin Watson, Regional Director, AI & Industries at monday.com.

Australian managers are becoming ‘bot bosses’ on the job

Work allocation is already changing inside Australian businesses. Only 14% of directors and senior managers default to assigning most tasks to people and using AI as a secondary tool. Almost half (45%) preserve work requiring human judgement and offload the remainder to AI, while more than one in four (26%) assume AI will handle most execution and employees will act primarily as editors and strategists.

These agents can receive instructions, access company information and act across connected systems, yet they sit outside conventional headcount, performance management and succession planning. That leaves managers responsible for a growing digital workforce without established rules for ownership, authority or offboarding:

• • Nearly half (44%) fear agents won’t receive sufficient human oversight or control, exposing a gap between what the technology can do and who is checking it
  • Almost one-in-three (31%) fear being held accountable for an agent’s work, leaving managers responsible for digital workers they may not fully control
  • 84% believe good management requires different skills in the age of AI, pointing to a capability gap as leaders take charge of mixed human and agent teams
    
    

The shift is colliding with Australia’s emerging guardrails for agent use. The Australian Signals Directorate has cautioned organisations against granting agents broad system access or decision-making authority without human review, while the Joint Select Committee on Artificial Intelligence is examining the effects of AI on productivity, employment, skills and regulation ahead of its report at the end of this month. 

The more agents touch the work, the harder it is to review the human

Agent sprawl is also making individual contributions harder to trace. A strategy deck may now pass between an employee, several AI tools and an AI-enabled manager before it is finished. The final product may be faster and more polished, but reveal less about who framed the problem, challenged a weak answer or caught the critical mistake. That loss of visibility is already forcing Australian employers to rethink performance reviews:

• • Four-in-five (80)% of surveyed Australian organisations have changed how they evaluate employees since adopting AI
  • 37% have significantly changed their evaluation criteria, the highest result across the seven markets
  • Nearly six-in-ten (59%) say an employee’s original contribution has become harder to evaluate 
  • 40% use AI to draft performance reviews
    
    

“Agent sprawl makes work harder to coordinate and talent harder to see. When several agents help research, draft and assess the same output, polish and speed stop being reliable evidence of ability. Businesses risk rewarding the best-presented work rather than the strongest judgement, while overlooking the people who set the direction, challenged the answer, caught the mistake and took responsibility for the result. Unless performance systems capture those decisions, companies will lose sight of where human capability actually sits,” said Watson.