--- format: "aidr-story-markdown/v1" id: "04117300ff14c04fd96b93086378e00732ab92fd0811173a7a88680df3139e95" canonical_url: "https://aidr.today/04117300?lang=en" title: "Anthropic Defaults Paid Claude Plans to Auto Mode" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-10-22T21:38:24.000Z" category: "Products" topics: ["anthropic","safety","agent"] source_urls: ["https://marketbrief.now/ai/anthropic-defaults-paid-claude-plans-to-auto-mode-0efeebac","https://huggingnews.com/ai/anthropic-defaults-paid-claude-plans-to-auto-mode-0efeebac"] summary: "Anthropic is switching Claude Pro, Max, and Team subscribers to a default auto mode that uses an internal safety classifier to block risky actions instead of requiring manual confirmation for each step. Testing showed auto mode caught 89% of dangerous commands compared to 14% via manual approval. Users can still switch back to manual permissioning at any time." --- # Anthropic Defaults Paid Claude Plans to Auto Mode > [Open the canonical story]() **Published:** 2026-10-22T21:38:24.000Z **Category:** Products **Topics:** anthropic, safety, agent ## Summary Anthropic is switching Claude Pro, Max, and Team subscribers to a default auto mode that uses an internal safety classifier to block risky actions instead of requiring manual confirmation for each step\. Testing showed auto mode caught 89% of dangerous commands compared to 14% via manual approval\. Users can still switch back to manual permissioning at any time\. ## Sources - [Story source]() - [Story source]()