Microsoft launches Scout, a personal AI agent based on the OpenClaw framework
Microsoft Corporation has introduced Scout, a new always-on, artificial intelligence-based assistant for the Microsoft 365 ecosystem.
Microsoft Corporation has introduced Scout, a new always-on, artificial intelligence-based assistant for the Microsoft 365 ecosystem.
Microsoft Corporation has introduced Scout, a new always-on, artificial intelligence-based assistant for the Microsoft 365 ecosystem.
As TechCrunch reports, Scout was conceived as a flexible assistant with a permanent identity and its own style, working in close tandem with the user. The main feature of the agent is its ability to learn and adapt to the individual work habits of a particular person based on their constant feedback. Users can even give their own names to their digital assistants.
Scout Project Vice President Omar Shahine explained the philosophy behind the new tool, emphasizing the importance of personalization and long-term engagement.
«We all have our own interesting quirks at work, and people encode these patterns into memories and skills that are stored in their agent. Then the agent becomes more capable, understands you better, gains more autonomy and demonstrates the ability to make judgments,» noted Omar Shahine.
Scout currently runs in the cloud, but it integrates with desktops and web browsers. This makes it easy to connect it to corporate email, calendars, and other internal systems. The assistant comes with a basic set of skills — for example, for scheduling meetings and structuring plans for the day. However, the developers expect that the most value will come from the unique skills that users create for the assistant themselves during their daily routine.
Microsoft has paid special attention to the security of autonomous work. Since at the beginning of the year the OpenClaw open agent repeatedly demonstrated unpredictable behavior (in particular, cases of chaotic sending of emails from the researcher’s mailbox were recorded), Scout implemented a «policy compliance system». It continuously monitors the agent’s work according to specified security criteria and conducts a detailed audit of all its actions.
Currently, the new AI assistant is only available to participants in the Microsoft Frontier experimental program who have an active subscription to GitHub Copilot.



