Anthropic Folds Cowork Into a Single Claude Experience Across Plans

Anthropic announced on September 16, 2026 that Claude Cowork and Claude chat are merging into one Claude, with the unified experience rolling out to Pro and Max plans over the coming weeks and two new tools, Claude Docs and Claude Slides, launching in beta.

In the merged experience, there is no mode to choose. Users describe what they need, and Claude decides which tool to use, answering a quick question or taking on larger work such as research, reports, spreadsheets, and presentations and handing it back as finished files that can be edited. Those files include spreadsheets with working formulas and presentations that open in PowerPoint.

Anthropic said it built Cowork as a separate place for bigger work and Claude Design for visual work, and that users told the company the frustrating part was deciding where a task belonged, because work started in one product did not carry into the other. The company said Claude can now figure out what a task needs, so Cowork and Design capabilities are available from any conversation with the user’s existing context, skills, and connectors.

Claude Cowork began as a research preview for Max plans on macOS on January 12, 2026 and expanded to Pro plans on January 16, 2026. It reached general availability on macOS and Windows on April 9, 2026; Claude Design followed on April 17, 2026 as an Anthropic Labs product for visual outputs such as designs, prototypes, slides, and one-pagers. Cowork arrived on web and mobile on July 7, 2026, when chat and Cowork began sharing one home, and memory began working across chat and Cowork in the cloud on August 25, 2026, according to Anthropic’s release notes.

Claude Docs, Claude Slides, and Claude Design

Claude Docs and Claude Slides launched the same day, and Claude Design now works inside conversations. All three are in beta on paid plans, and Enterprise admins choose when to turn them on for their organizations. Users can ask for a document and write it together with Claude, or ask for a presentation and have Claude draft the slides; results can be edited directly, presented from Claude, or downloaded as PowerPoint or PDF. Claude Design used on its own keeps working as before.

According to Anthropic’s Help Center documentation, Claude Design makes on-brand visuals and mockups, Claude Slides makes presentations, and Claude Docs produces living documents written with Claude and a team, shareable by link or export. Anything made with the three tools lives at one shareable link that opens on a phone, where users can select and move an element or tell Claude what to change.

Tasks, Connectors, and Permissions

For multi-step work, users describe the outcome and Claude works through the steps on its own: searching the web, reading files, running code, and assembling the results. Several tasks can run at the same time, and more involved tasks keep running in the cloud even if the user closes a laptop or leaves the page, while tasks that use files or apps on the user’s computer require the Claude Desktop app to be open.

Connected apps such as Google Drive, Gmail, Microsoft 365, and Slack work while Claude carries out a task. Claude can also browse the web using the browser built into Claude Desktop or the user’s own Chrome browser through Claude in Chrome, and a computer use capability that lets Claude click, type, and navigate the screen is in beta on Pro and Max plans. Users can schedule recurring tasks that run in the cloud, check on or redirect a task from the mobile app, and teach Claude skills or add plugins.

A permission setting in the message box controls how independently Claude works. Manual, the default, asks the user before Claude takes actions; Auto lets Claude keep working, with automated safety checks running before each action. The setting applies to the whole conversation and can be changed at any time, and users can stop or redirect Claude while it works. The announcement tells users they keep the final say.

The announcement walks through a weekly-report example: a user asks Claude what moved in the pipeline, then tells it to write the report the usual way, flag anything that slipped, and put the highlights into five slides for a leadership meeting, with the report and slides coming out of the same conversation. In the same post, Andrew Keller, a senior economist, said Claude could pull up his legal research database, retrieve and read the cases, identify other cases he might need, and store them in a folder for his review.

Rollout, Migration, and Stated Limitations

The new experience is rolling out first to existing and new users on Pro and Max plans in the Claude app on web, desktop, and mobile over the coming weeks, with nothing for users to turn on. The rollout is staged, so accounts on the same plan will see the change at different times; if the message box still shows separate Chat and Cowork options, the account does not have the new experience yet. Once an account moves over, it cannot switch back to the separate options.

Existing Cowork tasks, projects, connectors, skills, artifacts, and files carry over and open as before, and the model picker and Code tab remain where they were. Cowork tasks now live together with chats in Recents, and Cowork’s Global instructions setting has become part of Instructions for Claude in Settings. There is no web search toggle; Claude searches the web when it helps, and research is reached by typing /deep-research or choosing Research from the plus menu.

Anthropic lists several current limitations: Add from GitHub is not supported, branching a conversation from an earlier point is unavailable, incognito chats open in the previous experience without file creation or code execution, search does not include older Cowork tasks, and Dispatch is not available to new users. All activity counts toward a plan’s usage limits, longer agentic tasks generally use more than a quick question, and usage may be measured slightly differently for accounts that have the new experience during the rollout.

Anthropic said Team and Free plans will follow soon, and that Enterprise admins will hear from the company at least 30 days before anything changes for their organizations.