Anthropic's Claude Auto Mode Automates Coding with AI-Driven Safeguards

• Anthropic introduces "auto mode" for its AI assistant Claude, enabling independent coding tasks.
• The feature, a research preview, reduces manual approval by allowing Claude to evaluate task safety.
• Auto mode reviews each action for risks like unintended behavior or prompt injection attacks, blocking unsafe actions.
• It builds on previous tools like Claude Code Review and Dispatch, shifting decision-making to the AI.
• Auto mode will roll out to enterprise and API users, recommended for isolated environments due to its research preview status.