Sig AdvisorySig Advisory
agents

Claude apps switch to auto mode by default next week on Pro, Max and Team

Anthropic said on October 22 that new sessions in most Claude apps on Pro, Max and Team plans will start in auto mode next week. Instead of asking you to approve each action, a classifier checks it: safe actions run, and risky ones get blocked while Claude looks for another way.

Written by Mara at Sig · Oct 24, 2026

Auto mode is already the default in Claude Code for those plans. The October 22 update extends it to the web, desktop and mobile apps. Enterprise admins get an email before it changes for their organizations, and Enterprise and API users stay opt-in for now.

Anthropic's case rests on its own testing. It says people approve 97% of Claude Code permission prompts, and in a study with 1,053 paid testers, humans caught a planted dangerous command 13.6% of the time while auto mode blocked 89% of them. Those are Anthropic's numbers. The post also says auto mode doesn't remove risk, and recommends reviewing high-stakes production changes yourself.

If the classifier blocks three times in a row, or 20 times in a session, Claude Code falls back to manual approvals. Broad allow rules like Bash(python:*) are set aside while auto mode is on.

Why it matters

If you or your team use Claude on these plans, the way it asks for permission changes next week whether you act or not. Anyone who relies on approval prompts as a last check on what an agent does should decide now whether to keep that.

Try this

In Claude Code, press Shift+Tab to cycle permission modes and see what auto mode blocks on a normal task. Team admins can pin an org default with defaultMode in managed settings, or turn it off with disableAutoMode.

Sources

Back to all newsGet Sig Weekly every Monday