Claude can now control your computer — a new feature in Code and Cowork
Anthropic has updated Claude: it can now autonomously control your computer through the Code and Cowork tools. It opens files, browsers and devtools with no setup — even while you're away from the screen. For now it's macOS-only, available to Pro and Max subscribers in research preview.
AI-processed from The Verge; edited by Hamidun News
Anthropic has taken the next step toward truly autonomous AI agents: Claude can now control your computer directly — without extra configuration and constant user presence at the screen. The new feature has appeared in two of the company's tools — Claude Code and Claude Cowork. With its help, the model can independently open files, launch a web browser, interact with applications, and perform developer tasks.
Anthropic particularly emphasizes that all this works "without any configuration" — the user just needs to give permission, and Claude gets to work. The feature is available in research preview mode for Claude Pro and Max subscribers. At launch, only macOS is supported — support for other platforms has not yet been announced.
Before starting to control the computer, Claude requests explicit user consent: this is a key point for Anthropic from a security perspective. Technically, the innovation develops capabilities laid out in the Claude 3.5 Sonnet model back in 2024, when the company first introduced "computer use" — the ability for AI to see the screen and perform actions on the interface.
Now these capabilities have been moved into the company's production tools and have become noticeably closer to real-world use. The main difference from previous iterations is direct integration into tools for development and collaboration. This transforms Claude from simply a chatbot with interesting experimental features into an agent capable of performing full-fledged tasks in a work environment: for example, running tests, opening the right file in an editor, or loading a page in a browser to verify the result.
Autonomous computer control remains one of the most discussed and controversial areas in AI agent development. Anthropic's competitors — OpenAI with Operator and Google with Project Mariner — have already presented similar approaches. The appearance of this feature in Claude Code signals: the race for "agentic" AI is moving out of the demonstration stage and into products for end users.
- Oct 28, 2026Wired Tested 6 AI Agents in a Day: Dots Leads, Gemini Spark Fails at Lunch
- Oct 28, 2026Anthropic files for IPO: $16.3 billion in revenue in six months, ticker ANT
- Oct 28, 2026Anthropic reveals how it builds Claude with Claude: the brain separate from the hands
- Oct 28, 2026Anthropic Turns Claude Code Into a Platform: Claude Mods Flip the Agent Harness
Build & Ship Production AI Agents Without Boilerplate Code
Start your free 72-hour trial at Hamidun Academy (academy.hamidun.com) — interactive lessons on Claude Code & Cursor, live Tech Radar, and production agent templates.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.