_1777294237270_124456903.png&w=48&q=75){kind=small}
Claude introduces AI-managed Gmail messages
Anthropic announced the addition of a new feature to their smart assistant, Cloud, which allows sending emails directly through Gmail upon user request. The assistant can automatically compose and reply to messages, with a prompt for user confirmation before sending as a security measure. However, this option can be disabled to enable direct operation without human review, which carries technical and regulatory risks due to the potential for errors or incorrect messages reaching the wrong recipient. Cloud's capabilities also include searching emails, organizing the inbox, and managing message threads, but it is limited to reading email metadata on attachments and cannot access their internal content. Other companies like Google and Microsoft adopt a more cautious approach, where AI is only used to draft messages, leaving the final sending decision to the user.
Notice: This Is an AI-Generated Summary
Comments (0)