If you’ve ever opened an AI product and then spent ninety seconds wondering whether you needed the chat box, the “work” mode, or the design canvas, Anthropic heard you. On Wednesday the company said it’s folding Claude’s chat and Cowork features into a single interface, and pulling its visual tool Claude Design into that same home screen.
The motivation is almost charmingly human: users were frustrated by having to choose the right tool before they could start the actual work. Claude, Anthropic says, will increasingly decide which capabilities a task needs and deploy them without making you play product manager for your own prompt.
It’s the kind of UX confession every AI lab eventually makes—feature sprawl is fun for demos and exhausting for deadlines.
Docs, Slides, and the export button that matters
Alongside the unification, Anthropic is launching Claude Docs and Claude Slides in beta. The pitch is simple: stay in one conversation and come out the other side with documents, presentations, and visual designs instead of a pile of markdown that still needs a human to dress it for meetings.
Export options land where office reality lives. Documents can head to Google Docs or Microsoft Word; presentations can leave as PowerPoint or PDF. That sounds mundane until you remember how many “AI work” tools still strand you in a proprietary preview that doesn’t survive the email thread.
In other words: fewer screenshots of the chatbot, more files your boss can actually open.
The enterprise arms race, summer edition
Anthropic’s timing isn’t subtle. OpenAI launched ChatGPT Work in July, stitching its chatbot together with the Codex coding stack. Both companies are racing for enterprise seats and for non-coders who want AI-powered work tools without becoming prompt engineers full-time. Both are also widely understood to be IPO-bound giants, which means every interface tweak doubles as a product roadmap and a market narrative.
Unifying surfaces is also a quiet admission that “suite of AI apps” can lose to “one assistant that doesn’t make you context-switch.” Cowork, chat, and Design each had a reason to exist. Users still voted with their confusion.
Who gets it first
The rollout starts with Pro and Max subscribers across web, desktop, and mobile over the coming weeks, then expands to Team and Free plans later. That’s the familiar freemium staircase: prove the workflow on paying power users, then democratize once the edges stop sparking.
For Team admins, the interesting question isn’t whether Claude can draft a slide deck—it can—but whether the unified agent keeps permissions, brand templates, and review trails intact when it auto-selects tools. Anthropic didn’t turn this announcement into a compliance white paper, so enterprises will do what enterprises do: pilot it on the marketing wiki and panic later.
Why this is more than a tab cleanup
AI products spent two years multiplying modes: research mode, agent mode, canvas mode, “deep think,” and whatever else fit on a dropdown. Each mode taught a model to behave differently—and taught users to second-guess themselves. Collapsing those modes into one conversational surface is a bet that routing should be the model’s job, not yours.
If Claude’s auto-routing is sharp, the update will feel like an OS that finally stopped asking which printer driver you meant. If it’s fuzzy, people will invent folklore about secret keywords that “force” Design or Docs anyway. Either outcome will be very online.
For now, Anthropic’s message is refreshingly unglamorous: stop making humans choose the toolbox. Make the toolbox show up when needed, export to the formats offices already use, and get out of the way. In a market obsessed with frontier benchmarks, shipping fewer buttons might be the most competitive move of the week.
Source: Reuters — Anthropic to fold Claude AI features into one interface, launches document tools