Anthropic is turning Claude Code’s auto mode on by default starting Aug 14
Anthropic announced that auto mode will become the default setting for Pro, Max, and Team accounts starting August 14, reducing manual oversight.
Stock photo for illustration only, not from the actual event
- Anthropic makes auto mode the default for Pro, Max, and Team accounts
- The change takes effect on August 14, cutting down manual prompt approvals
- Testing showed auto mode outperformed human reviewers in catching harmful actions
- New safety features include prompt injection screening and customizable hard deny rules
Programming with Claude Code will soon require even less human oversight, as Anthropic announced plans to make its auto mode the default setting for Pro, Max, and Team accounts, with the rollout scheduled to begin on August 14.
The company originally unveiled a test version of auto mode back in March, pitching the feature as a practical method to balance speed and control. As Anthropic explained in an announcement on Friday, when Claude Code operates in auto mode, it proceeds with tasks automatically instead of displaying pop-up prompts that require human approval at every single step—unless an action is determined to be irreversible, destructive, or directed outside the user's local environment.
Stock photo for illustration only, not from the actual event
The company also noted that during testing, auto mode proved to be safer than manual code reviews. In a formal study involving 1,053 paid testers, auto mode successfully caught 89% of harmful actions, whereas human reviewers only caught 13.6%. The disparity may stem from the habituality of manual reviews, as users tend to approve a staggering 97% of permission prompts in Claude Code without deep inspection.
Shifting auto mode to the default setting highlights Anthropic's focus on mitigating prompt fatigue among developers who frequently face repetitive approval dialogs. While automating these decisions accelerates workflows, it places heavy reliance on robust guardrails to ensure that autonomous code execution does not trigger unintended destructive outcomes within sensitive development environments.
"The team and I use Auto mode exclusively, and have been for many months. I couldn’t imagine going back to permission prompts!"
Boris Cherny, Head of Claude Code
Stock photo for illustration only, not from the actual event
Source: TechCrunch
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment