Claude’s new Cowork feature threatens Gemini’s Workspace advantage — and puts dozens of startups at risk

Claude logo on phone
(Image credit: Shutterstock)

Earlier this week, Anthropic made waves by announcing a launch into healthcare, just days after OpenAI revealed ChatGPT Health.

Now, the company just announced Cowork, a new research preview that turns Claude from a conversational assistant into something closer to an autonomous digital coworker — one that can directly read, create, edit and organize files on your computer, plan multi-step tasks and carry them out with minimal supervision.

While this new feature is currently now only available to Claude Max subscribers on macOS, this move is huge. Cowork dramatically expands what Claude can do beyond chat. And while it’s positioned as a “simpler” way to work with Claude, its implications are anything but small, especially for small businesses and startups.

What Cowork actually does

Organize your desktop files with Cowork - YouTube Organize your desktop files with Cowork - YouTube
Watch On

Cowork gives Claude access to a folder on your computer that you explicitly choose. From there, Claude can:

  • Read and reorganize files
  • Rename and sort documents
  • Create spreadsheets from screenshots
  • Draft reports from scattered notes
  • Generate documents and presentations
  • Work through multiple queued tasks in parallel

Instead of waiting for a prompt-response loop, Claude creates a plan and executes it step by step, checking in as it goes. If you’ve used Claude Code, the experience will feel familiar — but Cowork removes the need for coding entirely.

Pair it with browser access, and Claude can complete tasks that involve web research, downloads or form-filling. Pair it with existing connectors and it can pull in external data automatically.

In practice, Anthropic is promising Claude Cowork feels like assigning work to a colleague and coming back later to review progress.

Why this is a big deal for non-developers

Claude app

(Image credit: Future)

Until now, most “agentic” AI tools — systems that can take actions rather than just respond — have been developer-focused, complex or fragmented across multiple apps. Now, Cowork lowers that barrier dramatically.

You don’t need to:

  • Manually re-upload context
  • Copy-paste outputs into the right format
  • Babysit every step of a task

Users can simply queue instructions, add feedback mid-process and let Claude handle the execution. That alone puts Cowork in a different category from traditional AI chat tools.

The startup problem no one wants to talk about

Claude 4

(Image credit: NPowell/Flux-Kontext)

Here’s the part that should make founders nervous. Cowork directly overlaps with dozens of startups built around:

  • AI file organization
  • AI-powered document generation
  • AI agents for admin work
  • “Second brain” productivity tools
  • Workflow automation for non-technical users
  • AI-powered spreadsheet creation
  • Research and reporting assistants

Many of these startups exist specifically to bridge the gap between chat-based AI and real work on your computer. Cowork erases much of that gap in one move — and it does so inside a tool millions of people already use.

Instead of launching a nich feature, Anthropic has potentially absorbed entire product categories into its core platform. For startups, the risk isn’t that Cowork does everything better today. It’s that:

  • Cowork will improve rapidly
  • It’s deeply integrated into Claude’s core models
  • It doesn’t require users to adopt a separate workflow
  • It benefits from Anthropic’s scale, trust and safety infrastructure

In other words, even Cowork has made standalone tools feel redundant.

Control, safety and real risks

Claude

(Image credit: Shutterstock)

Anthropic is careful to emphasize guardrails. Claude can only see folders you explicitly grant access to, and it asks before taking significant actions. But there are real risks, including:

  • Accidental deletion of local files
  • Misinterpreting instructions
  • Prompt injection attacks via web content

Anthropic has always put safety first, building defenses against these issues, but agent safety remains an active area of development — and Cowork may be the first time many users experience an AI tool with this level of autonomy.

The takeaway

This is a significant moment signaling a broad shift in AI, especially as the race heats up every day. For big tech, it's no longer about who has the best chatbot, but which one truly owns the workspace.

By turning Claude into an AI that can actually do the work — not just suggest it — Anthropic is moving into territory once dominated by productivity apps, automation platforms and niche AI tools.

Cowork is available now as a research preview for Claude Max subscribers on macOS, with Windows support and additional safety features planned in future updates.


Google News

Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds.


More from Tom's Guide

Amanda Caswell
AI Editor

Amanda Caswell is an award-winning journalist, bestselling YA author, and one of today’s leading voices in AI and technology. A celebrated contributor to various news outlets, her sharp insights and relatable storytelling have earned her a loyal readership. Amanda’s work has been recognized with prestigious honors, including outstanding contribution to media.

Known for her ability to bring clarity to even the most complex topics, Amanda seamlessly blends innovation and creativity, inspiring readers to embrace the power of AI and emerging technologies. As a certified prompt engineer, she continues to push the boundaries of how humans and AI can work together.

Beyond her journalism career, Amanda is a long-distance runner and mom of three. She lives in New Jersey.

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.