Anthropic makes auto mode Claude Code default August 14
Anthropic announced that starting August 14, auto mode becomes the default permission mode in Claude Code for Pro, Max, and Team users.
- Auto mode's classifier caught 89% of dangerous commands in testing, versus 14% for manual approval.
- Classifier overhead no longer counts toward usage limits on Pro, Max, and Team plans as of today; Enterprise and API users get the same later.
- In a study of 1,053 paid testers, humans caught dangerous commands 13.6% of the time, dropping to about 5% after 50 prompts.
- Users can switch modes with Shift+Tab, and admins can pin defaultMode or disable auto mode in managed settings.
Claude Code users on Pro, Max, and Team plans will no longer approve each command manually, letting multi-hour tasks run unattended with fewer interruptions.
Sources
Read this as text
Back to the AI news