AI agents are becoming increasingly prevalent at work, going beyond generating content to taking action, following processes, and making decisions within set boundaries. Organizations began exploring agentic AI much like they did with generative AI – starting with experimentation. But now, many have moved beyond that early phase, as recent research reported that 90% of organizations have progressed beyond solely experimenting with AI agents1. And we’re starting to see this impact adoption metrics across industries and job levels – we found that 2 in 3 employees report using agentic AI at least once a week. And this was after being provided with an aided definition and examples to differentiate it from generative AI, showing that this technology is quickly becoming part of everyday workflows.
Chapter 1 of our 2025 Agentic Teaming & Trust Research Report dives into the emergence of agentic AI at work. In this chapter, we explore how momentum and interest is building around AI agents and how we’re starting to see early, multi-dimensional impact. We also focus on what phases organizations are in on their way to having established agentic AI transformation strategies.
Not only are employees beginning to leverage agents more frequently, but momentum around agents is also growing as well. More than 70% of employees see the value of integrating agentic AI at work and are motivated to do so. And for those that are already using agents at work, they are starting to perceive real impact of using them – in greater productivity, decision-making, and work quality. This multi-dimensional impact of agentic AI is starting to show the ways agents can support individual employee work (catch chapter 4 for team-based impact!).
As employees’ knowledge of various AI agents is growing, their preference on what type of agent they want to leverage/manage is getting clearer. 90% of our participants demonstrated a preference for a single type of agent, even though they were able to say they wanted all three types of agents presented. These agent types included a Versatile Agent, a Domain Expert Agent, and an Automated Process Agent, each offering distinct capabilities. But this clarity didn’t drive a “winner” between the agent types – the preferences were split.
As organizations look to transform with agentic AI, understanding employee use cases will be critical. 67% of leaders in our research reported being in the “planning” or “implementing” stage of re-engineering workflows to fully leverage the capabilities of agentic AI and only 16% have fully established talent strategies to help upskill employees to manage, monitor, and collaborate with agentic AI.
When it comes to making agentic AI real, lean into the emerging excitement from your employees to capture the momentum and drive positive change. Make sure every employee – from interns to execs – have the support, resources, and opportunity to integrate agentic AI into their processes. Start by identifying high-impact workflows, investing in upskilling programs, and creating cross-functional agentic AI task forces. Getting everyone on board and making sure transformation is not just top-down will push your organization further into a agentic AI future.
Want to explore the full story?
Download the PDF - Chapter 1: Ready for Agents
The Agentic Teaming & Trust Study was conducted by the Microsoft People Science team utilizing an Online Panel Vendor, commissioned by Microsoft, with 1,800 full-time employees across nine markets between June 11, 2025 and July 7, 2025. This survey was 12 minutes in length and conducted online. Global results have been aggregated across all responses to provide a total or average. Each sample was representative of business leaders across regions, ages, and industries (i.e., Construction, Financial & Professional Services, Retail, Food, & Beverage, Healthcare, Media & Entertainment, Technology, Transportation, Travel, & Hospitality). Each sample included specific parameters on company size (i.e., organizations with 1,000+ employees) and job level (i.e., business leaders/business decision makers, those in mid- to upper job levels such as C-level executive, VP or director, Manager). The overall sampling error rate is 2.31 percent at the 95 percent level of confidence. Markets surveyed include Brazil, China, France, Germany, India, Japan, Mexico, United Kingdom, and the United States. Findings represent aggregated responses and may not reflect all organizations or industries.
References
Microsoft