Study Questions Productivity Boost from AI Agents

A new study from work management platform Asana is challenging the narrative that AI “agents” are dramatically transforming workplace productivity. The research suggests that while agentic AI systems — tools designed to autonomously complete tasks — are attracting heavy investment and attention, their real-world impact on efficiency and output remains mixed.

According to Asana’s findings, most companies experimenting with AI agents report only modest gains. Employees said the tools often struggle to understand context, integrate across platforms, or execute complex multi-step workflows without human supervision. Instead of eliminating repetitive work, the study found that AI agents frequently require extensive oversight, reducing the time savings they were supposed to deliver.

The survey also revealed a disconnect between expectations and outcomes. Nearly two-thirds of executives believe AI agents will transform their operations within the next two years, yet only a fraction of frontline workers report measurable improvements to their daily tasks so far. Asana noted that without clear goals, training, and strong data governance, agentic AI can generate more noise than value.

However, the study stops short of calling AI agents a failure. It points to best practices that can unlock their potential, including narrowing their scope to high-volume, low-risk tasks, providing structured data, and integrating them into existing workflows rather than replacing them outright.

Asana’s report underscores that AI agents are not a magic bullet. Organizations hoping for instant ROI may be disappointed unless they invest in thoughtful implementation and continuous monitoring.