Microsoft Explores Integration of OpenClaw-Agents into 365 Copilot for Enterprise Markets

Microsoft is testing the integration of features inspired by the OpenClaw platform into its enterprise AI assistant 365 Copilot. This is reported by The Information, citing sources within the company. According to journalists, efforts are directed at enabling Copilot to operate autonomously around the clock — executing tasks on behalf of the user without constant manual management.

Microsoft's Corporate Vice President Omar Shahin confirmed the company's interest in this direction, stating that Microsoft is exploring the potential of OpenClaw-like technologies in an enterprise context. This is the first official acknowledgment from the company of specific interest in the platform. Previously, Microsoft discussed agentic AI in general terms, without naming specific open projects as a reference point.

OpenClaw is an open-source platform for creating AI agents that run locally on a user's device. Unlike cloud-based AI assistants, OpenClaw agents have direct access to the file system, applications, browser, and local computer data — without necessarily transmitting information to external servers. This makes the platform particularly attractive to the enterprise market, where data security and regulatory compliance are critical.

Companies fear transmitting confidential documents and business correspondence to third-party cloud services — and local agents could be a solution to this problem. OpenClaw's popularity surged in early 2026. Developers worldwide began building agents based on it to automate routine tasks: scheduling management, incoming email processing, document analysis, form filling, and much more.

In March 2026, the community held its first major conference — ClawCon in New York — which became a landmark event: from a niche developer tool, OpenClaw transformed into a full-fledged technology movement with thousands of participants. Today, 365 Copilot operates on a query-response model: the user asks a question or gives a command — the assistant responds. This is fundamentally different from an agentic model, in which AI acts proactively.

Integration of an OpenClaw-like architecture would allow Copilot to independently process incoming mail and respond to standard requests, create tasks in a scheduler based on meeting outcomes, prepare weekly reports by a set time, track deadlines and warn of risks — all without manual activation by an employee. This is the scenario Microsoft describes as 24/7 autonomous operation. The race for agentic AI systems for the enterprise market is already in full swing.

Google introduced Agentspace — a platform for enterprise agents with Workspace integration. Salesforce is developing Agentforce. Amazon and Oracle announced their own solutions in this space.

Microsoft, with one of the world's largest enterprise customer bases through the Microsoft 365 package, is in a strong position to scale agentic technologies. If internal tests prove successful, the updated Copilot capabilities could appear in releases soon — the company does not specify timelines, but the market is not waiting. For now, it is a matter of research, not a ready product.

Nevertheless, Microsoft's confirmation of interest in local AI agent technologies speaks to an important shift: enterprise AI is moving from the role of an assistant that responds to questions to the role of an independent executor. For business, this promises the liberation of working time previously spent on routine tasks. At the same time, it raises new questions: how to control autonomous AI actions, how to set boundaries for its authority, and how to ensure transparency of decisions it makes in a corporate environment.