Anthropic announced on September 16 the merger of Claude chat and Cowork, while introducing Claude Docs and Claude Slides. Users can ask questions, assign tasks, write documents, and create presentations within the same conversation, instead of manually selecting a separate area for each type of work.
According to the company, the unified interface will begin rolling out to Pro and Max subscribers on web, desktop, and mobile over the next few weeks. This is a phased rollout, meaning not all accounts will receive the update immediately on the day of the announcement.
Claude handles tool selection
Previously, Cowork was designated for larger tasks, while Claude Design served design and visual work. Anthropic stated that users found it challenging to decide where to start, while work completed in one area did not automatically carry over to another.
The new interface brings Cowork and Design capabilities into standard conversations. Claude automatically determines which tool to use based on the user’s prompt, context, skills, and connected services.
An example provided by Anthropic is asking Claude to write a weekly report and then turning key points into a presentation. Both deliverables originate from the same conversation, eliminating the need for users to copy all content into another tool to start over. According to the company, work can continue after users close their computers and can be tracked from their phones.
This is primarily a change in how the product is used. The announcement is not the launch of a new model, but rather Anthropic bundling existing capabilities and new tools into a more accessible workflow.
Documents and presentations still require human finishing touches
Claude Docs allows users to co-author and edit documents with AI. With Slides, users can request presentation generation, make direct edits, present directly within Claude, or download them as PowerPoint or PDF. Products created with Docs, Slides, and Design can be shared via a link and opened on mobile phones.
These tools are currently in beta on paid plans; Enterprise administrators decide when to enable them. For the unified interface, Anthropic noted that Team and Free plans will be supported later, while Enterprise customers will receive at least 30 days’ advance notice regarding relevant changes.
A notable detail is that Claude defaults to asking before taking actions. Users can choose to let the AI continue working and only prompt again when closer review is needed. Convenience thus comes with decisions regarding the level of autonomy granted to the tool.
If it works as described, the most obvious benefit is reducing window-switching and repeating prompts. However, the ability to quickly generate a report does not replace checking data, sources, and content before sending it to others. The quality of the final output remains the real-world test for this new way of working.