The race to build powerful AI agents that can take over complex tasks on your computer is intensifying — and Microsoft is now working to bring that capability to its enterprise customers in a more controlled, secure way. Microsoft is testing ways to integrate OpenClaw-like features into its existing Microsoft 365 Copilot tool. The new features would be geared toward enterprise customers, with better security controls than the famously risky open source OpenClaw agent.
The move signals that Microsoft sees local AI agents — tools that run directly on a user's machine and autonomously complete tasks — as the next major frontier in enterprise productivity software.
What Is OpenClaw and Why Does It Matter?
For those unfamiliar, OpenClaw is a tool that runs locally on a user's computer and can create agents to perform tasks on behalf of the person. It has become one of the most talked-about open source AI projects in recent months, attracting a devoted user base drawn to its ability to automate complex, multi-step workflows without sending data to the cloud.
However, its open source, locally-run nature has also made it a security concern. A Meta AI security researcher previously documented an OpenClaw agent that ran amok on her inbox — a vivid illustration of what can go wrong when autonomous agents operate without robust guardrails. Microsoft's enterprise version aims to solve exactly that problem.
Where Claude Fits In
Notably, Anthropic's Claude is already part of Microsoft's agentic strategy. Microsoft has tapped Anthropic's Claude to power Cowork, after it partnered with the AI lab late last year. Microsoft added Claude as an option available for Cowork. While OpenClaw can work with multiple models, Claude remains the model of choice for many users of the open source project.
The fact that Claude is central to both the open source OpenClaw ecosystem and Microsoft's enterprise Copilot efforts underscores Anthropic's growing influence across the agentic AI landscape.
Microsoft's Growing Arsenal of AI Agents
The new OpenClaw-inspired agent would not exist in isolation — it would join a rapidly expanding suite of agentic tools Microsoft has been rolling out. In March, Microsoft announced Copilot Cowork, which is designed to take actions in Microsoft 365 apps rather than just providing search results or chat in a separate work pane. Cowork is powered by its own "Work IQ" technology, an intelligence layer that personalizes Cowork for the user across Microsoft 365 apps.
In February, Microsoft also introduced Copilot Tasks, another agent designed to complete tasks, released at the time in preview. The marketing materials suggested it was geared toward prosumers, with tasks ranging from organizing email to organizing travel and appointments — tasks that go beyond Microsoft's Office suite. Both Cowork and Tasks run in the cloud, not locally.
Always On, Always Working
What makes the new Claw agent distinctive is its intended behavior. Microsoft told The Information that one of the main features of the agent is that it would essentially be a version of 365 Copilot that is always working, able to take actions at any time. The idea is an agent that can complete multi-step tasks over long periods.
That "always on" capability — persistently running in the background and autonomously handling tasks without constant user prompting — represents a meaningful leap beyond what current Copilot tools offer. For enterprise users managing complex workflows across multiple Microsoft apps, this could be genuinely transformative.
The Mac Mini Problem Microsoft Wants to Solve
There's also a competitive dimension to Microsoft's effort that goes beyond enterprise features. While the open source project OpenClaw can run on Windows machines, the Mac Mini has been the go-to platform for OpenClaw users — so much so that the small, affordable, cube-shaped Mac Mini desktops have suddenly been selling like hotcakes.
That's a problem for Microsoft. If OpenClaw's popularity is driving hardware sales to Apple, building a compelling Windows-native alternative becomes a strategic priority — not just a product decision.
Build Conference Reveal Expected
Microsoft hasn't fully detailed what this new agent will look like, but a public debut appears to be coming soon. The company is expected to show off the new Claw agent — or an upgraded version of one of its existing Claw-like tools — at its Microsoft Build conference in June.
For enterprise IT teams watching the agentic AI space, Build 2026 just became a date to circle on the calendar.
