Claude no longer distinguishes between Chat and Cowork, as Anthropic begins to make Agent mode invisible.

动察 Beating AI News Flash: Anthropic has merged Claude Chat and Cowork into a single entry point. No more choosing between "chat" and "work" first — just hand your request to Claude. It will decide on its own whether to answer a question or take on a complex task. Projects, Skills, Connectors, and context from Cowork are all preserved.

Interestingly, Cowork was originally spun off by Anthropic on its own initiative. When it launched in January this year, it was positioned as "Claude Code for the rest of your work," specifically handling multi-step tasks like reports and documents. But once users actually started using it, they still had to figure out whether a task should go to Chat or Cowork, and work across the two entry points was hard to carry on naturally. Now Anthropic has proactively removed that layer of choice.

Claude has also added Docs and Slides, allowing users to generate and edit documents and presentations directly within a conversation, and Claude Design has been integrated into the same entry point. Once a task is handed to Claude, it can continue processing even if you shut down your computer, and users can check progress on their phones.

Anthropic started with Cowork to push Claude from a chat tool into a working Agent, and now it has hidden the Agent mode itself. Users don't need to care whether it's Chat, Cowork, Docs, or Slides behind the scenes — they just tell Claude what they want.