Editorial illustration for Microsoft's Autopilot AI Agent Gets Its Own Identity and Workspace
Microsoft Splits Copilot Into Three AI Agents
Microsoft's Autopilot AI Agent Gets Its Own Identity and Workspace
Microsoft rolled out a redesigned Copilot app on Tuesday, splitting the assistant into three distinct pieces: Home, Code and Autopilot. The change marks a shift away from Copilot as a single chat window toward something closer to an operating layer for work, one that spans documents, meetings, emails and the agents doing tasks in the background.
Autopilot gets the most attention here. Microsoft describes it as a persistent, proactive agent, one with its own identity and workspace, that keeps working on tasks after a person has logged off or moved on to something else. That's a departure from how Copilot has functioned so far, where the assistant mostly waited for a prompt before doing anything.
The rollout isn't instant. Home and Code start reaching Microsoft's Frontier program in the coming weeks, while Autopilot moves to private preview at the end of the month. Microsoft is pairing the release with new FinOps for AI tools, aimed at helping businesses track spending as they hand more work over to agents like this one.
The Copilot app now has three new capabilities: Home is your new starting point, where Chat and Cowork come together — and with Office in Copilot, the full power of Word, Excel and PowerPoint is now built into the experience. Code lets everyone build their own solutions with the tools to run them safely, and it’s powered by the same underlying technology as GitHub Copilot.
Why this matters
Giving Autopilot its own identity, memory and workspace inside a tenant is a bigger move than the "chat, meet office productivity" framing suggests. Microsoft is treating an AI agent as something closer to a directory object than a feature toggle, complete with @mentions in Teams and Outlook like it's a coworker with a badge. For developers and IT teams, that raises real questions we'll be watching: how permissions and access controls get scoped to an agent's "identity," what "Microsoft IQ" actually pulls from to model an organization, and who audits an agent's memory when it's persistent across chats, channels and documents.
For founders building on Copilot, this is a signal that Microsoft wants agents to be first-class citizens in the org chart, not bolt-ons, which changes how you'd architect integrations. The upside is agents that understand context without constant re-prompting. The risk is an agent with standing memory and workspace access becoming a new attack surface or compliance headache nobody's fully mapped yet.
Related Reading
Browse all Microsoft news →
AI Tools & Apps Claude gains shared context in Excel, PowerPoint; Microsoft adds Copilot Cowork
AI Tools & Apps OpenAI Developer: AI Agent "Swarms" Waste Tokens, Create "Coordination Tax
AI Tools & Apps LG's recent webOS update adds Microsoft Copilot app, now removable
Related Topic Microsoft adds ’vibe working’ to Word and Excel; Copilot Agent Mode now default
Business & Startups
Related Topic Nadella: AI success needs intense usage, Cloud hits USD 54.5B, Azure up 40%
Business & Startups
From Archives Meta’s Muse Tracks Your Email and Bank Data While You Work
From Archives Microsoft Patches Vulnerabilities Amid AI Bug-Hunting Surge
From Archives Microsoft and OpenAI Workers Warn AI Is Harming the Web
Research & Benchmarks
From Archives Microsoft internal docs reveal AI scraping details in NY Times lawsuit
Business & Startups
Common Questions Answered
How does Microsoft's redesigned Copilot app split the assistant into different components?
Microsoft split Copilot into three distinct pieces: Home, Code, and Autopilot. Home serves as the new starting point where Chat and Cowork come together with full Office integration including Word, Excel, and PowerPoint. Code enables users to build their own solutions with tools to run them safely, powered by the same technology as GitHub Copilot.
What makes Autopilot different from previous Copilot implementations?
Autopilot is designed as a persistent, proactive agent with its own identity and workspace that keeps work organized across documents, meetings, emails, and background tasks. Unlike the single chat window approach, Autopilot functions more like an operating layer for work and is treated as a directory object similar to a coworker with a badge in Teams and Outlook.
How does Autopilot's identity and workspace affect permissions and access controls?
By giving Autopilot its own identity, memory, and workspace inside a tenant, Microsoft is raising important questions about how permissions and access controls get scoped to an agent's identity. This approach treats AI agents as directory objects rather than simple feature toggles, which requires developers and IT teams to reconsider how they manage agent-level access and authorization.
What is the significance of Autopilot having @mentions capability in Teams and Outlook?
Autopilot's ability to receive @mentions in Teams and Outlook positions it as a coworker-like entity within the workplace communication ecosystem. This integration suggests that AI agents are becoming integrated into standard workplace collaboration tools and workflows, fundamentally changing how teams interact with AI assistance.
Further Reading
- Product Hunt - AI Tools - Product Hunt
- There's An AI For That - TAAFT
