When does auto mode become the default?

Auto mode becomes the default setting for new Claude Code sessions starting August 14th for Pro, Max, and Team plans.

What did Anthropic's testing show auto mode could block?

In a test of 1,053 paid users, auto mode blocked 89% of harmful actions that participants were prompted to approve midway through sessions, compared to only 13.6% of humans who refused the same harmful action.

Did third parties independently verify Anthropic's security claims?

Anthropic commissioned Trajectory Labs to test prompt injection scenarios; Trajectory Labs reported that none of 720 attack attempts succeeded against Claude Fable 5, Opus 5, or Sonnet 5 running auto mode. However, broader independent security community confirmation has not yet occurred.

Large Language ModelsAI Coding AssistantsSimon Willison's WeblogAug 13, 2026Aug 13, 2026, 10:01 JST

Claude Code auto mode becomes default for Pro, Max, Team users

Send on LINE

3 Key Points

What happened
Anthropic is making auto mode the default setting for new Claude Code sessions across Pro, Max, and Team plans starting August 14th. The company claims auto mode blocks 89% of harmful actions in a test of 1,053 paid users, where participants were shown a dangerous command midway through sessions.
Why it matters
Auto mode reduces confirmation fatigue—asking humans to approve every step leads to unsafe behavior. In the same test, only 13.6% of humans refused the harmful action, suggesting humans approve risky commands at high rates under repeated prompts. Anthropic is betting auto mode is safer than human judgment for routine decisions.
What to watch
Anthropic commissioned Trajectory Labs to test 72 indirect prompt injection scenarios against Claude Fable 5, Opus 5, and Sonnet 5 running auto mode as of July 17th 2026, reporting none of the 720 attack attempts succeeded. However, the security community remains skeptical—the setting rolls out August 14th, and independent verification of these claims is still pending.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

FAQ

When does auto mode become the default?: Auto mode becomes the default setting for new Claude Code sessions starting August 14th for Pro, Max, and Team plans.
What did Anthropic's testing show auto mode could block?: In a test of 1,053 paid users, auto mode blocked 89% of harmful actions that participants were prompted to approve midway through sessions, compared to only 13.6% of humans who refused the same harmful action.
Did third parties independently verify Anthropic's security claims?: Anthropic commissioned Trajectory Labs to test prompt injection scenarios; Trajectory Labs reported that none of 720 attack attempts succeeded against Claude Fable 5, Opus 5, or Sonnet 5 running auto mode. However, broader independent security community confirmation has not yet occurred.

Simon Willison's WeblogRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.