Editorial illustration for Anthropic releases Claude Code, Cowork for macOS in research preview
Claude Code Brings AI Automation to macOS Desktops
Anthropic releases Claude Code, Cowork for macOS in research preview
Anthropic just rolled out a new research preview that lets its Claude models interact directly with a computer. Dubbed Claude Code and Cowork, the tools are designed for macOS users who subscribe to Claude Pro or Max, promising a hands‑on way to automate tasks without writing scripts. While the company has hinted at broader autonomous features in earlier releases, this is the first time the AI can actually control a desktop environment.
The move follows a series of incremental upgrades that have expanded Claude’s ability to reason and generate code, but the real test will be whether the system can reliably execute commands in a live OS. For developers and power users, the prospect of an LLM that can open apps, edit files or run terminal commands could shave minutes off routine workflows. Yet Anthropic tempers expectations, noting that the preview is still confined to a single platform and a specific subscriber tier.
The research preview is currently limited to macOS devices. These new capabilities are available as a research preview for Claude Pro and Max subscribers, and computer usage is limited to macOS "for now," according to Anthropic. The feature builds on autonomous capabilities that were introduced on C
The research preview is currently limited to macOS devices. These new capabilities are available as a research preview for Claude Pro and Max subscribers, and computer usage is limited to macOS "for now," according to Anthropic. The feature builds on autonomous capabilities that were introduced on Claude's 3.5 Sonnet AI model in 2024, but now brings those to the chatbot's Code and Cowork AI agents for programmers. To access the feature, the Claude desktop app must be running on a supported macOS device and paired with the chatbot's mobile app.
Anthropic’s latest research preview lets Claude Code and Cowork take the reins on a Mac, opening files, browsing the web and even firing up development tools without any user‑side configuration. For now, the capability is confined to macOS devices and is only accessible to Claude Pro and Max subscribers, meaning the broader user base won’t see it yet. What’s striking is the claim that the AI can act “even when you’re away from your computer,” suggesting a level of autonomy that goes beyond simple command‑following.
Yet the announcement offers no detail on how security, privacy or error handling are managed, leaving those concerns largely unaddressed. The feature builds on earlier autonomous capabilities Anthropic introduced, but whether it will scale beyond the current sandbox remains uncertain. If the preview proves reliable, it could become a useful productivity aid; if not, the promise may outpace practical utility.
Only further testing will reveal how well the AI balances convenience with control.
Further Reading
- Cowork Preview Lets Claude Handle Office Tasks on Your Mac, With Risks Attached - PureAI
- Anthropic previews Cowork as Claude expands beyond coding tasks - EdTech Innovation Hub
- First impressions of Claude Cowork, Anthropic's general agent - Simon Willison's Blog
- Anthropic Expands Claude's 'Computer Agent' Tools Beyond ... - ADT Mag
Related Reading
Browse all Anthropic news →
LLMs & Generative AI Ant Group unveils Ring-1T, first open-source trillion-parameter reasoning model
LLMs & Generative AI ChatGPT Health Event Shows AI Modernizing Dev Workflows, GitLab Unveils Plans
LLMs & Generative AI Gen AI app sessions up fivefold, downloads jump 778% as ChatGPT leads traffic
Related Topic Google launches AI chips with 4× boost, lands Anthropic multibillion deal
Business & Startups
Related Topic Anthropic finds strict anti-hacking prompts increase AI sabotage and lying
Research & Benchmarks
From Archives Zuckerberg creates personal 'CEO agent' as Anthropic's Claude gets remote control
From Archives Google NewFront 2026: Gemini advantage shows integrated G‑MP campaigns add value
From Archives Anthropic launches Claude Code Channels for Telegram and Discord messaging
From Archives Cursor launches Composer 2, outperforms Claude Opus 4.6, lags GPT‑5.4
Common Questions Answered
What specific capabilities do Claude Code and Cowork offer for macOS users?
Claude Code and Cowork enable AI interaction directly with a computer, allowing the AI to open files, browse the web, and launch development tools without user configuration. These tools are currently available as a research preview exclusively for Claude Pro and Max subscribers on macOS devices.
How does the new Claude research preview differ from previous autonomous AI capabilities?
Unlike earlier hints at autonomous features, this research preview allows Claude to actually control a desktop environment for the first time. The capabilities build on the autonomous features introduced with Claude's 3.5 Sonnet AI model in 2024, now integrated into the Code and Cowork AI agents for programmers.
What are the current limitations of the Claude Code and Cowork research preview?
The research preview is currently restricted to macOS devices and only accessible to Claude Pro and Max subscribers. The AI can perform tasks like opening files and browsing the web, with Anthropic suggesting the AI can even operate when the user is away from the computer.