AI technology is advancing rapidly and now it seems to be reducing the role of humans further. Anthropic has introduced a new Auto Mode in its AI model Claude, which can take decisions on its own without the user's permission every time. This feature is specifically designed to reduce constant interference during coding. However, this has also raised questions about the increasing independence of AI and whether it is safe or a matter of concern.
Anthropic says that the new Auto Mode enables Claude to complete many tasks without asking for permission from the user every time. Especially during coding, the need to give permissions again and again is reduced. In this feature, AI first checks the risk level of every task and takes decisions on that basis. Safe tasks are completed automatically, while on any suspicious or risky task the AI stops and asks permission from the user. This increases the speed of work and also reduces frequent interruptions.
Auto Mode is said to be safer than the earlier dangerously-skip-permissions option. In that old feature, AI used to work directly without any investigation, which increased the risk. But in the new system, Claude identifies potential threats and tries to prevent them. Tasks like deleting large amounts of files, running dangerous code or leaking sensitive data are automatically blocked. In this way this feature tries to strike a balance between convenience and security.
Currently, Anthropic has released Auto Mode as a research preview for Claude Team users. Soon it will also be made available for API and enterprise users. This feature supports Claude Opus 4.6 and Sonnet 4.6 models. Users can turn it on or off from the settings of their desktop app. However, the company has advised to always use it in a sandbox or isolated environment, so that any potential risks can be reduced.
