Saturday, April 25, 2026

“Government Releases 2026 Income Tax Return Forms”

The approaching income tax season signals the...

“Ford Blasts Carney’s China EV Deal, Citing Lack of Communication”

Ontario Premier Doug Ford expressed his frustration...

Glenn Maxwell Excluded from Cricket Australia Contracts

Australian veteran all-rounder Glenn Maxwell has been...

“Anthropic Expands Claude AI Computer Use to Windows”

Technology"Anthropic Expands Claude AI Computer Use to Windows"

Anthropic recently introduced the computer use feature for Claude Cowork and Claude Code, enabling the AI to independently perform various tasks on a computer, including app creation and testing. Initially available only on macOS, this functionality has now been extended to Windows.

According to a post by the official Claude account, computer use in Claude Cowork and Claude Code Desktop is now accessible on Windows. This feature empowers the AI to open applications, navigate interfaces, conduct tests, identify bugs, and resolve them autonomously. Anthropic claims that the entire process can be completed with a single prompt.

To utilize the Claude computer use feature on Windows, users must be subscribed to Claude Pro or Claude Max and download the relevant apps. No additional setup or terminal commands are required as Claude leverages existing system resources for task execution. Additionally, users can remotely assign tasks to Claude using the Dispatch feature via their smartphones.

For security purposes, users need to grant app access to Claude per session, limiting the AI’s control to only approved applications during that specific session. The system maintains a lock to ensure single-session control by Claude and allows users to halt any action instantly using the Esc key or terminal commands.

In response to user feedback about faster depletion of Claude usage limits, Anthropic clarified that users were not charged extra. Lydia Hallie from Anthropic explained that peak-hour limits became stricter and 1M-context sessions were enhanced, causing the perceived increase in token usage. She assured users that no overcharging had occurred and recommended using the Claude Sonnet 4.6 model for Pro subscribers, as it offers better performance compared to Opus.

Check out our other content

Check out other tags:

Most Popular Articles