Anthropic announced that Claude Code‘s auto mode will become the default for Pro, Max, and Team accounts starting August 14.

Auto mode, first tested in March, reduces human oversight by automatically proceeding with actions unless they are deemed irreversible, destructive, or outside the user’s environment.

The company cited a study with 1,053 paid testers where auto mode caught 89% of harmful actions, compared to 13.6% for manual review.

Anthropic noted that manual review often becomes habitual, with users approving 97% of permission prompts in Claude Code.

Claude Code Head Boris Cherny stated on X that the team has used auto mode exclusively for months.

Additionally, new safety features such as prompt injection screening and customizable hard deny rules have been introduced to prevent data exfiltration and other risks.