In just three months, Microsoft customers have built over 1MN custom agents across Copilot Studio and SharePoint.
That's up by a remarkable 130 percent compared to the previous quarter.
Additionally, over 230,000 organizations have begun using Copilot Studio to start creating these custom agents. That includes 90 percent of the Fortune 500.
Copilot Studio provides a low-/no-code interface that allows anyone to create agents using natural language prompts.
These agents can integrate with data sources through pre-built or custom plugins and be deployed across diverse enterprise environments.
Satya Nadella, Chairman & CEO of Microsoft, shared the rising success of Copilot Studio during the tech giant's Q1 2025 earnings call.
The CEO explained that recent upgrades to the agent builder were key to these impressive results.
"With deep reasoning and agent flows in Copilot Studio, customers can build agents that perform more complex tasks and also handle deterministic scenarios like document processing and financial approvals," he said.
However, he also underscored the impact of the new computer use feature for Copilot Studio, which launched on April 15, 2025. Nadella continued:
[Customers] can now build computer-use agents that take action on the UI across desktop and web apps.
More specifically, this capability allows businesses to build AI agents capable of handling tasks on websites and desktop apps without needing an API.
These agents can automate data entry, market trend analysis, and various other cross-system processes.
Microsoft also ensures the tool operates on its secure hosted infrastructure, eliminating the need for businesses to manage servers.
Elsewhere, Microsoft has enhanced its consumer Copilot by enabling it to book restaurant reservations, purchase event tickets, and shop online in the background.
That further beckons the future of machine-to-machine customer service.
Microsoft's AI Agents for Customer Service and Sales
Via the Copilot studio, customer service teams can build AI agents for all types of use cases.

