Microsoft is preparing another major reset for Copilot, according to an internal memo seen by The Information. The reported plan points to a clear shift: Copilot is being pushed away from experimental AI features and toward tools that can help people complete practical work.
The overhaul is reportedly set to release in August. It would bring consumer and enterprise Copilot apps together into a single app, add AI coding tools, and introduce new AI agents called "AutoPilot" that can handle background tasks such as scheduling and email summaries. Customers would pay extra for those added features.
A single Copilot app with more work built in
The most important product change is consolidation. Instead of separating consumer and enterprise apps, Microsoft plans to merge them into one app. That matters because it suggests the company wants Copilot to feel less like a set of disconnected AI experiments and more like a central place where different kinds of work can happen.
The reported app would include AI coding tools, putting software development inside the same broader Copilot push. That places coding alongside everyday productivity tasks, which is consistent with Microsoft’s stated emphasis on outcomes rather than AI novelty.
The new AutoPilot agents are also central to the plan. According to the memo, they would work in the background on tasks like scheduling and email summaries. That is a different kind of AI product from a chatbot that only responds when a user asks a question. Background agents imply a more persistent role inside workflows, where the software is expected to take action or prepare useful output without requiring every step to be manually prompted.
Microsoft also plans to charge customers extra for the added features. That makes the new Copilot strategy a product and business test at the same time: the app has to be useful enough that customers see value in paying more for it.
What Microsoft is removing from the Copilot strategy
The memo also describes a cleanup of features that did not fit the new direction. Executive Vice President Jacob Andreou wrote that the team "stripped out what wasn't working," including Copilot Podcasts and Copilot Labs.
Those removals show how Microsoft is narrowing Copilot’s purpose. The memo says Copilot should focus on "real work" instead of chasing intelligence "for intelligence's sake." It should be "optimized for outcomes," and the app has to "earn the right to exist," Andreou wrote.
That language is unusually direct for a product reset. It frames Copilot not as a showcase for model capability, but as software that must justify its place in a user’s day. In plain terms, the new measure of success is not whether Copilot can demonstrate impressive AI behavior. It is whether Copilot helps people complete tasks that matter.
The AI super app race is getting crowded
Microsoft is not alone in trying to turn AI assistants into broader work platforms. The source article says Anthropic with Claude Code and OpenAI with Codex are working on similar "super apps." In that context, the Copilot overhaul looks like part of a wider race to build AI products that sit closer to daily workflows.
The common idea is simple: an AI assistant becomes more valuable when it can connect multiple work functions in one place. For Microsoft, that means a Copilot app that can cover enterprise use, consumer use, coding, scheduling, and email summaries. For customers, the question is whether those pieces feel like one useful system or just more features in the same interface.
The mention of Claude Code and Codex also shows how important coding tools have become in this category. Coding is not described in the source as a separate side project. It is part of the same move toward outcome-focused AI software, where the product is judged by what it can help users produce or complete.
Why workplace deployment matters
Microsoft also announced a new company focused on rolling out AI inside businesses. The plan is for Microsoft engineers to work directly inside departments and help build AI into workflows.
That detail is important because it highlights a limit of the standalone chatbot model. The source article describes it as an admission that a chatbot alone delivers limited value, or at least value that is hard to measure. If AI is not connected to the way teams actually work, its impact can be difficult to prove.
Working inside departments is a more hands-on approach. It suggests that Microsoft sees adoption as an implementation problem, not just a software distribution problem. A company can have access to AI tools and still struggle to make them useful in routine work. Embedding engineers into departments is one way to bridge that gap.
The bigger test for Copilot
The Copilot reset comes as Microsoft and other AI companies still have to justify their billions in AI spending. That pressure helps explain the memo’s emphasis on practical results, paid features, and measurable work.
For Copilot, the challenge is now sharper. It must compete with other AI super app efforts, give users a reason to adopt one combined app, and prove that AutoPilot agents and AI coding tools are worth paying for. The reported August overhaul is therefore more than a redesign. It is a test of whether Microsoft can turn Copilot into a product that earns a lasting role in everyday work.