Starting August 14, Anthropic will make the Claude Code "auto mode" the default setting for Pro, Max, and Team accounts. The change is intended to reduce the need for step-by-step human approvals and speed up code generation workflows.
Since a test release in March, Anthropic has presented auto mode as a way to balance speed with user control. In a Friday announcement the company explained that when Claude Code operates in auto mode, it will proceed with actions unless those actions are classified as "irreversible, destructive, or aimed outside your environment," rather than prompting the user for approval at every step.
Safety results and internal use
Anthropic says that in testing auto mode proved safer than manual review. In a study involving 1,053 paid testers, auto mode detected 89% of harmful actions, while human review caught only 13.6%. The company notes that manual review can become habitual: according to the study users approved 97% of permission prompts in Claude Code.
Boris Cherny, Head of Claude Code, wrote on X: “The team and I use Auto mode exclusively, and have been for many months. I couldn’t imagine going back to permission prompts!”
Additional protections
Anthropic also reported adding further safety measures such as prompt-injection screening and configurable hard deny rules designed to prevent incidents like data exfiltration.
The switch to making auto mode the default for paid tiers takes effect on August 14, and Anthropic says it will continue to monitor and refine the feature.
