Anthropic is bringing Claude Cowork and its regular chat experience together into a single workspace, making Claude less dependent on users deciding how a task should be handled before they start it.

The company announced the change on September 16, saying users can now begin with a simple question or hand Claude a much larger project from the same conversation. Instead of choosing between Claude’s traditional chat and Cowork for more involved work, Claude can determine what the task requires and continue working from there.

The unified experience is rolling out first to Pro and Max subscribers across Claude’s web, desktop, and mobile apps, with Team and Free plans set to follow. Enterprise customers will receive advance notice before the change reaches their organizations.

Claude no longer makes users choose between chat and Cowork

Cowork was originally designed as a separate environment for larger projects and more involved computer-based work. That separation created a practical problem: users had to decide where a task belonged, and work started in one environment did not necessarily carry over naturally to another.

Anthropic says it built the new experience around removing that decision.

With Cowork and chat now combined, Claude can use the capabilities previously associated with Cowork directly from a conversation. That includes the context, skills, and connectors already available to the user.

In practice, a task can start as an ordinary conversation and become something much more substantial without requiring the user to switch modes.

For example, Anthropic describes a scenario where someone asks Claude to analyze what changed in a company’s pipeline over the previous week, write the results in the team’s usual format, flag anything that slipped, and turn the highlights into a five-slide presentation for a leadership meeting.

Claude can work through those steps from the same conversation, producing both the report and presentation. The user can then review the work, make changes, leave comments for Claude, and share the resulting files.

Anthropic also says users can check on work from their phones while Claude continues working, including after they have closed their laptops.

Claude Docs and Claude Slides arrive alongside the merger

The Cowork consolidation is also part of a broader expansion of Claude’s workspace capabilities.

Anthropic has introduced Claude Docs and Claude Slides, while Claude Design is now available directly inside conversations. Users can ask Claude to create a document or presentation as part of the same interaction rather than treating those tasks as separate destinations.

Documents and presentations can be edited directly, presented from Claude, or downloaded as PowerPoint or PDF files.

Anthropic says Claude Docs, Claude Slides, and the integrated Claude Design experience are currently in beta for paid plans. Enterprise administrators will control when the features are enabled for their organizations.

Anything created with Claude Design, Claude Slides, or Claude Docs can also live behind a single shareable link. Users can open that work from their phone, move individual elements, or tell Claude what they want changed.

More control over how Claude works

The unified workspace does not mean Claude has to operate completely autonomously.

Anthropic says Claude asks for approval before taking an action by default. Users can instead enable a mode where Claude continues working and only checks in when something requires closer attention.

That gives users a choice in how much supervision they want to provide while keeping the final say with them.

The approach is particularly relevant for longer-running tasks. Rather than requiring users to remain inside a conversation and repeatedly provide instructions, Claude can continue working on a project and return with a completed result or a request for clarification.

Anthropic is also bringing over existing Cowork work. Users who have already been using Cowork will retain their chats, projects, artifacts, connectors, and skills when they move to the unified Claude experience.

A shift from AI chat to AI workspace

The bigger change here is less about adding another Claude feature and more about changing how Anthropic wants people to interact with the product.

Traditional AI chat interfaces are built around a question-and-answer model: ask something, receive an answer, and continue the conversation. Cowork introduced a different model in which Claude could take on larger pieces of work.

Anthropic is now combining those two approaches.

A user does not necessarily need to know whether a task is “chat work” or “Cowork work.” They can simply describe what needs to get done and let Claude determine how much work is required.

That also makes the new document, presentation, and design tools more useful because they are no longer isolated destinations. A conversation can naturally move from research to writing, from writing to presentation creation, and from presentation creation to editing without forcing the user to restart the workflow elsewhere.

For people using Claude for recurring work, Anthropic is also adding scheduling to the workflow described in its announcement. In its example, a weekly report can be scheduled for every Monday, allowing Claude to begin preparing it without the user having to initiate the task each time.

The rollout is happening gradually. Pro and Max users are receiving the unified experience across web, desktop, and mobile over the coming weeks, while Team and Free plans will follow. Enterprise administrators will receive at least 30 days’ notice before changes are made for their organizations.

Anthropic’s broader direction is becoming clear: Claude is moving beyond being a place where users ask an AI questions and toward becoming a workspace where the AI can participate in completing the work itself.